An unprecedented number of cancer patients now survive, for a total of 10 million in the United States. But long-term cardiac or pulmonary toxicity can be a problem for breast cancer patients who received anthracyclines, radiotherapy, or trastuzumab. Since many of these late effects are asymptomatic, how and when should they be identified?
Two recent reports in the Journal of Clinical Oncology highlight the dangers of anthracyclines, specifically the irony that longer life means a greater likelihood of cardiac problems.
One report, by a group of researchers at the Department of Breast Medical Oncology at the University of Texas M. D. Anderson Cancer Center and the University of Texas Medical Branch at Galveston, evaluated the rates and predictors of congestive heart failure in women over age 65 who had received adjuvant anthracycline therapy (JCO 2007;25:3808-3815).
Data were compiled from the Surveillance, Epidemiology and End Results (SEER) Medicare database between 1992 and 2002, and the results showed significantly higher rates of congestive heart failure compared with the rates in women who had received nonanthracyline therapy or no chemotherapy at all.
Figure. Joseph Carve...Image Tools
The other report, in the same issue, was a clinical evidence review by an American Society of Clinical Oncology expert panel on the ongoing care of adult cancer survivors related to cardiac and pulmonary late effects (JCO 2007;25:3991-4008).
ASCO wanted to develop guidelines for long-term survivors of anthracyclines, so a group of us started to work on it, and we did this big literature review, said the lead author and member of the writing committee, Joseph Carver, MD, Chief of Staff at the Abramson Cancer Center at the University of Pennsylvania and Clinical Professor of Cardiology.
We wanted to see if there was consensus about this, added Charles L. Shapiro, MD, Professor of Internal Medicine and Director of Breast Medicine Oncology at Ohio State University. We also reviewed the cardiac and pulmonary effects of radiation, chemotherapy-related pulmonary toxicity, and trastuzumab-related cardiotoxicity.
But when we presented our report to the ASCO board, said Dr. Carver, They said there wasn't enough hard data to create practice guidelines.
Instead, the board asked the panel to review the literature without creating formal guidelines. To have been included in the published review, a study must have examined a screening intervention for cardiovascular or pulmonary disease; the study population had to have been cancer survivors who had received chemotherapy, radiotherapy, and/or trastuzumab; and the outcome had to have been the effect of screening on cardiovascular or pulmonary morbidity, quality of life, and overall survival.
Preliminary searches identified 4,805 abstracts, but applying the criteria winnowed the articles to 364, although few were designed to address toxicity screening in asymptomatic survivors.
Reasons for Concern
Anthracycline-based therapy works. It has reduced the annual breast cancer death rate by 38% for women younger than 50 and by 20% for those age 50 to 69. Women over 70 have a risk reduction similar to younger women.
But anthracyclines have a downside: cardiotoxicity, the most serious of which is congestive heart failure. Although the problem is well known, there are limited data about the long-term consequences. For example, it is widely believed that a cumulative dose of 500 mg/m2 is the upper limit for relative safety, but cardiomyopathy can and does occur at doses as low as 300 mg/m2.
Other issues are that in children, cardiac problems increase with time, and that in adults, no one knows the long-term incidence of congestive heart failure because it has never been measured.
Data about patients over 65 are even more murky and may be far worse than suspected. Fewer older women participate in clinical trials compared with younger patients, older patients tend to have more comorbidity, and they are not assiduously followed.
Doxorubicin is the most problematic of the anthracyclines, causing anatomic or functional abnormalities of the heart and lungs. Cardiac dysfunction following trastuzumab rarely causes death and is usually reversible after discontinuation.
Radiation-induced cardiac disease occurs in 10% to 30% of patients five to 10 years after treatment, although that is decreasing with the advent of modern techniques for radiotherapy.
What the SEER Data Showed
First author of the study by the Texas researchers, first presented as an abstract at the 2006 ASCO Annual Meeting, was Mary C. Pinder, MD. The original study population included 43,338 women, of whom 4,712 (11%) received adjuvant anthracycline therapy; 10,096 made a Medicare claim for congestive heart failure.
Figure. Charles Shap...Image Tools
There were important differences in baseline characteristics: Patients who received any chemotherapy were younger and healthier, but with more advanced disease; anthracycline-treated patients were slightly younger and healthier than patients who received other drugs; black women were over-represented compared with their number in the general population; women with comorbidity were under-represented; and those treated with any chemotherapy had higher grade tumors and more estrogen receptor (ER)-negative disease.
Other findings:
* Despite being younger and healthier than women on other regimens, those age 66 to 70 who received anthracyclines were more likely to develop congestive heart failure: 19% at five years after treatment vs 18% on other chemotherapy and 15% for those not on chemotherapy. At 10 years, the difference was more pronounced: 38.4%, 32.5%, and 29%, respectively.
* For women age 71 to 80, there were no significant differences among the three groups, perhaps because they already had a higher baseline incidence of congestive heart failure. Or perhaps it was only the healthiest of this age group who received anthracyclines-at lower cumulative doses.
* Black patients had a 49% higher risk of congestive heart failure than whites did, probably because they had more advanced cancer.
* Preexisting coronary artery disease, emphysema, diabetes, hypertension, and peripheral vascular disease all predicted subsequent congestive heart failure for patients receiving anthracyclines.
What the ASCO Literature Review Showed
The literature review by the ASCO Cancer Survivorship Expert Panel-co-chaired by Dr. Shapiro and David J. Vaughn, MD, Medical Director of the Abramson Family Cancer Research Institute of the University of Pennsylvania, showed the following:
* The most important toxic effect of anthracyclines is abnormalities in left ventricular size and function.
* In children, the latency period may be as long as 25 years, and more than half will have echocardiogram or MUGA (multigate angiogram) abnormalities.
* The risk of cardiotoxicity was found to depend on total cumulative dose, age (under 18 or over 65 carries the highest risk), preexisting cardiac disease, concurrent mediastinal radiotherapy, combination chemotherapy, and length of survival.
* Trastuzumab-related cardiac damage rarely causes death, is not dose related, and is almost always reversible. However, it should not be used concurrently with doxorubicin.
* Strategies have been proposed for early detection of anthracycline cardiomyopathy: endocardial biopsy, enzyme testing, MUGA scan, exercise testing, and echocardiogram, but none can yet be considered standard.
* There is no evidence for the value of any treatment, although congestive heart failure may improve with beta blockers, angiotensin-converting enzyme inhibitors, diuretics, and nitrates.
* The heart is susceptible to radiation damage, especially in children treated for Hodgkin's disease and adults treated for breast cancer, head and neck cancer, lymphomas, and seminomas. Risk factors include concurrent anthracycline treatment, the location of the tumor near the heart, young age, baseline cardiac disease, total dose to the heart of more than 30 Gy, and fractionation of more than 2 Gy/day.
* Cardiac symptoms can appear up to 20 years after radiation therapy. The incidence is 10% to 30% by five to 10 years, but up to 88% of all patients have asymptomatic abnormalities of cardiac muscle.
More Prospective Studies Needed
Although the general risk of cardiotoxicity has been known almost since anthracyclines first came into use, the SEER data showed that more prospective studies are needed. To facilitate them and to obtain better data, the ASCO panel suggested using universal definitions and standardized tests of cardiac function as well as long-term follow-up of patients in controlled trastuzumab-anthracycline trials.
They are probably the best source of information on cardiac effects, Dr. Shapiro said.
He and his coauthors recommend that future research focus on critical areas: adoption of standard definitions of cardiac dysfunction, as well as standard tests to evaluate them; collection of incidence data in adults and children; and prospective long-term studies designed to determine the efficacy of screening.
Figure. Aman Buzdar,...Image Tools
Screening
Both JCO articles noted that no study has been designed to evaluate the utility and effectiveness of regular screening in asymptomatic patients.
Dr. Carver said that all survivors of anthracycline therapy should be screened for cardiac damage after a certain amount of time. At five years out, you see an increase in the incidence of left ventricular dysfunction, so you should start the screening then-echocardiogram and MUGA scan-and continue it at some interval afterwards.
He added that subclinical problems are much more common than people realize. We can intervene with medication, but we need to be vigilant.
Dr. Shapiro disagreed, however. Despite the exhaustive literature review, we [the ASCO panel] couldn't make a recommendation about screening because we lacked evidence that would support it, he said in an interview. A recommendation like this has enormous implications, and I feel it was prudent to refrain from making it. Creating guidelines, or even making recommendations, is not a trivial thing.
Of course, he continued, if a woman presents with signs and symptoms of CHF, she should be treated as medically indicated. But we're talking here about women who are well, and that's a totally different issue. Subclinical findings might never become a clinical problem, and even if we do detect them, we don't know if treatment will actually reduce cardiac death. So it does not make sense to screen healthy women when the only outcome is increased worry and expense.
Aman U. Buzdar, MD, Professor of Medicine in the Department of Breast Medical Oncology at M. D. Anderson, agreed with Dr. Shapiro: If a woman is asymptomatic and she doesn't need additional therapy with a potentially cardiotoxic drug, there is no need to evaluate cardiac function.
But what about the risk? The risk is very small, he replied. Close to zero after one or two years. But if the patient needs a second round of an anthracycline, then risk starts to increase.
Sandra M. Swain, MD, Medical Director at Washington (DC) Cancer Institute, said she doesn't routinely screen these patients unless they have symptoms. It might be a good idea, but it's terribly expensive. There are some trials going on now to compare screening vs not screening, so we'll see what they show.
'Renegade Recommendations'
Joseph Carver, MD, Chief of Staff of the Abramson Cancer Center and Director of the University of Pennsylvania Cardiology Fellow's Practice, described a set of what he called renegade recommendations that he, along with Andrea Ng, MD, of Dana-Farber Cancer Institute, Anna T. Meadows, MD, of Children's Hospital of Philadelphia, and David J. Vaughn, MD, of the University of Pennsylvania School of Medicine, developed when ASCO rejected the panel's proposed guidelines. They are recommendations, not guidelines, Dr. Carver emphasized, and are scheduled for publication later this year in the journal Disease Management.
Recommendations Applicable to All Long-Term Survivors:
* Stratify patients into high- and low-risk categories. Identify total dose of chemotherapy received, radiation dose and fields, and the results of baseline and treatment cardiac studies.
* Perform a directed history and physical examination: exercise tolerance, fatigue, chest pain, palpitations, and other signs of cardiac toxicity.
* Counsel patients about healthy lifestyle and behavior.
* Consider referral to a cardiologist or general internist if the patient has cardiac dysfunctions at a five-year screening test.
Recommendation Specific to Prior Anthracycline Treatment:
* Assess left ventricular function: an echocardiogram or radionuclide angiogram five years after completion of treatment and every five years thereafter.
Recommendations Specific to Prior Mediastinal Radiation Therapy:
* Assess ventricular function, pericardial status, valvular structure and function, and conduction system disease; screening echocardiogram at five years after completion of treatment and every five years thereafter.
* Assess risk of pericardial disease.
* Assess risk of coronary artery disease if the total dose was more than 35 Gy.
* Assess for coexisting abnormalities.
Recommendation for Testicular Cancer Survivors Treated with Cisplatin.
* Assess cardiovascular risk factors and the presence of vascular disease: Screening after completion of treatment and regularly thereafter.
Testing Is Not Recommended for These Survivors:
* Adults who are physically active with no limitation of functional capacity and whose cardiac risk factors are controlled.
* Asymptomatic patients treated with trastuzumab.
Accompanying Editorials
An editorial related to the study on congestive heart failure in older women with breast cancer treated with adjuvant anthracyclines by Mary C. Pinder, Zhigang Duan, James S. Goodwin, Gabriel Hortobagyi, and Sharon H. Giordano asks whether it is time to evaluate whether the benefits outweigh the risks of anthracyclines.
Because there are no criteria for the severity of congestive heart failure, the potential for misdiagnosis of mild or nonclassic symptoms is substantial, said Chia C. Portero, MD, of the National Cancer Institute and Sandra Swain, MD, of Washington (DC) Cancer Institute.
Also, they noted, cardiotoxicity sometimes occurred at a cumulative doxorubicin dose of 300 mg/m2, much lower than the recommended 500 mg/m2.
The link between radiation and CHF may be underestimated, but there is little information on the techniques used and the dose of radiation to the heart.
Drs. Portera and Swain made a surprising observation, they said, that anthracyclines do not increase the risk of congestive heart failure in women aged 71 to 80. This finding, they said, directly contradicts the results from other studies, and by focusing exclusively on CHF, the study might have missed an even broader scope of cardiovascular toxicity.
The editorial notes that other factors with potential for long-term cardiotoxicity should have been considered in the study, and that it would have been useful to have had a control group of women without breast cancer or receiving other chemotherapy.
However, they lauded the study for its valuable information on the anthracycline debate, which hinges on these questions: (1) Is it possible to minimize or eliminate cardiotoxicity from these drugs? (2) Which patients will benefit from anthracycline treatment? (3) Is it possible to replace anthracyclines with less cardiotoxic agents?
In an interview, Dr. Swain said that several approaches have been tried over the years: dexrazoxane, which has a poorer response rate (and in children, increased second malignancies); longer infusions (72 to 96 hours); anthracycline analogs; liposomal doxorubicin; and substituting a docetaxel-cyclophosphamide combination, which has just as good, if not better, efficacy as doxorubicin.
Dr. Buzdar suggested limiting the total cumulative dose of doxorubicin to 400 mg/m2 (or using epirubicin at 650 to 700 mg/m2).
In the editorial related to the ASCO panel report-Cancer Survivorship Research and Guidelines: Maybe the Cart Should be Beside the Horse-Craig C. Earle, MD, Associate Professor of Medicine in the Division of Population Sciences, Department of Medical Oncology, at Dana-Farber Cancer Institute, said, It is clear that creation of evidence-based guidelines for the care of adult cancer survivors is not currently feasible.
But why is this evidence so hard to obtain? Because cancer survivors are a heterogeneous population, and they are mostly elderly, so determining the late effects of the cancer itself as opposed to comorbid conditions can be difficult.
Moreover, endpoints of interest occur infrequently-that is, although the cumulative burden of late effects is significant, any one specific effect occurs relatively infrequently. Therefore, surveillance strategies or interventions would have to be applied to many patients over a long period of time, which is not practical.
Dr. Earle also said that the fact of so little guidance on how to care for survivors is in itself a challenge to research. For example, asymptomatic patients are not routinely screened, so there is no model of what to do for someone who has had an anthracycline or mediastinal irradiation. In these cases, physicians use their own judgment, which leads to wide variations in practice.
Why not treat these cardiac problems as if they had nothing to do with anthracyclines? Then it becomes a matter of individual care rather than large-scale research.
Dr. Earle explained that this is indeed the current state of the art. The issue is that there may be subclinical cardiac dysfunction, so later when there's a second insult to the heart, these patients may be more at risk than those without prior anthracycline exposure.
More research is necessary, he said, but in the meantime, guidelines can be as simple as advising use of history and physical examination to monitor for cardiomyopathy. It is possible, he added, to develop consensus guidelines that are informed by evidence rather than being strictly evidence based, which may be better than no guidance at all.
-MJF
© 2007 Lippincott Williams & Wilkins, Inc.