Field of cardio-oncology opens new pathways for consultation, collaboration, care
Over time, survival rates of patients with cancer have increased, but that has led to a quandary that was not well-characterized until recently: Many cancer treatments are cardiotoxic and, as a result, many cancer survivors develop and eventually die from some form of heart disease.
During the last few years, cardiologists have begun to take a greater role in the care of oncology patients. This has been occurring on several forefronts, including conducting research to learn more about how certain cancer therapies affect the heart and the CV system; helping to identify before initiation of cancer therapy who may be at risk for CVD; administering cardioprotective therapies that could mitigate the CV risk from oncology treatments; and refining treatment of CVD after cancer therapy. Some institutions have developed cardio-oncology programs to enable better collaboration between cardiologists and oncologists on these matters.
“The paradigm is changing,” Ana Barac, MD, PhD, FACC, director of the cardio-oncology program at MedStar Heart and Vascular Institute, Washington, D.C., medical director of the cardiac rehabilitation program at MedStar Washington Hospital Center and assistant professor of medicine at Georgetown University, told Cardiology Today. “The idea is: Let’s identify CV effects. Can we modify them? Can we make the treatments better and learn more through them?”
With the field of cardio-oncology in its infancy, there is still much to be learned, and there are few definitive trials or guidelines to show the way. However, what is known is that the ultimate goal for patients with cancer is to prolong overall survival and, to optimize that, CV effects have to be taken into account.
“We have recognized together that emerging chemotherapeutics in the oncology world, although very effective, have an undeniable CV risk. Unless we tackle both so-called ‘evils’ in a collaborative way, we cannot ensure the best results for our patients,” Gagan Sahni, MD, FACC, FACP, director of cardio-oncology and cardiology consult services at The Mount Sinai Hospital, told Cardiology Today.
Evidence of cardiotoxicity
According to current estimates, there are more than 14 million cancer survivors in the United States, and that number continues to grow every year. More than two-thirds of people with cancer are still alive 5 years after their diagnosis. “This means that they have time for potential CV side effects of chemotherapy agents to manifest,” Sahni said. “Also, these patients are becoming older, and as they get older their risk for CVD also becomes greater.”
Analyses of cohorts of patients with cancer have determined that CVD morbidity and mortality is a great risk in this population. For example, an analysis of more than 63,000 women with breast cancer from the Medicare Surveillance, Epidemiology and End Results (SEER) database revealed that CVD was a greater cause of death than breast cancer (15.9%; 95% CI, 15.6-16.2 vs. 15.1%; 95% CI, 14.8-15.4).
“What’s very interesting is that at the time of breast cancer diagnosis, only one-quarter of individuals had any CV risk factors,” W. Gregory Hundley, MD, FACC, FAHA, professor in internal medicine and radiology and medical director of CV imaging at Wake Forest Health Services, Winston-Salem, North Carolina, said in an interview. “That tells us two things. First, CV events are very important in individuals treated for cancer. Second, because the CV events occur after someone has been treated for cancer, it implies that the cancer treatments are somehow related to the subsequent CV event.”
As another example, a retrospective cohort study that matched 36,232 2-year survivors of adult-onset cancer with 73,545 controls without cancer determined that survivors of multiple myeloma (incidence rate ratio [IRR] = 1.7; P < .01), lung/bronchus carcinoma (IRR = 1.41; P < .01), non-Hodgkin’s lymphoma (IRR = 1.13; P < .01) and breast cancer (IRR = 1.13; P < .01) had higher risk for CVD than controls, particularly if they had at least two CV risk factors. Among those with cancer, 8-year overall survival was worse in those with CVD than in those without CVD (60% vs. 81%; P < .01).
Adverse CV effects from cancer treatments tend to occur more often in patients with pre-existing risk factors such as older age, hypertension and existing CAD, according to Sahni.
Research on Medicare and private-insurance databases has determined that hospital admissions for HF and other CV conditions “are more predominant after patients have started receiving treatments for cancer, implying that there’s something going on,” said Hundley, Imaging Section Editor of Cardiology Today.
Although some CV effects may happen immediately, most occur long after cancer treatment is finished. That could be 2 years, 5 years or maybe 10 years.
“Because we do so much better at getting many people cured of their cancer, they often have late-term cardiac side effects while they’re being followed during survivorship,” Randall Holcombe, MD, MBA, chief medical officer – cancer, The Mount Sinai Hospital System, told Cardiology Today. “It’s very important to understand what the long-term consequences of chemotherapy may be so that patients are followed appropriately and any cardiac issues are addressed.”
The evidence is not limited just to CV events. Using serum biomarkers and noninvasive imaging such as echocardiography and cardiac MRI, researchers have identified onset of subclinical CVD during and after cancer treatment with cardiotoxic agents. “The markers of subclinical disease that have been identified are associated downstream with future CV events,” Hundley said.
There is also evidence that childhood cancer survivors are more likely to develop metabolic health risk factors, including elevated BMI, hypertension and hypertransaminasemia, which can lead to premature adverse CV events.
Mechanisms and culprits
The harmful effects of cancer therapies on the CV system manifest in a variety of ways, experts told Cardiology Today.
The effects appear to occur in two general compartments, Hundley said. “One, the heart muscle structure itself, involving the myocytes and the extracellular matrix that supports those myocytes, and, two, the areas of the body responsible for fatigue or vascular events: the blood vessels, the microcirculation, and, I think, also the skeletal muscle.”
Many cancer treatments are designed to attack cells with a high rate of metabolism, so they may have an effect on the heart muscle cells, which have a very high rate of metabolism to enable the heart to beat, such as causing HF or cardiomyopathy, experts said.
In addition, some cancer treatments are designed to attack the vascular structure of tumors, which can have the unintended consequence of also attacking native vascular structures such as the CV system, and could lead to CAD or MI, they said.
Among the classes of chemotherapy drugs known to be related to adverse CV events include anthracyclines, HER-2–targeted therapies and tyrosine kinase inhibitors.
Anthracyclines, used to treat breast cancer, leukemia, lymphoma and solid tumors, have been associated with congestive HF. Women treated with anthracyclines for breast cancer “are often young when they present with cancer, and we see a significant late-term toxicity from the anthracycline treatment,” Holcombe said.
HER-2–targeted therapies such as trastuzumab (Herceptin, Genentech), which treat breast cancer, “are mechanistically associated with reversible or predominantly reversible effects on cardiac muscle,” Barac said. “It may not be associated with high frequency of HF per se, but it is still recommended to be used only in patients with normal ejection fraction.” Some research has suggested the agents are associated with as much as a fivefold increased risk for HF.
Some tyrosine kinase inhibitors, which are used to treat solid tumors, sarcomas and other cancers, “operate by increasing hypertension and, therefore, exposing the individual to increased risk for hypertension, cardiomyopathy and HF,” Barac said.
In a study published in February in JAMA Oncology, compared with imatinib (Gleevec, Novartis) treatment, patients assigned the tyrosine kinase inhibitors dasatinib (Sprycel, Bristol-Myers Squibb), nilotinib (Tasinga, Novartis) and ponatinib (Iclusig, Ariad) had an elevated risk for vascular occlusive events at 1 year (OR for dasatinib = 3.86; 95% CI, 1.33-11.18; OR for nilotinib = 3.42; 95% CI, 2.07-5.63; OR for ponatinib = 3.47; 95% CI, 1.23-9.78) without any improvement in overall survival.
Other cancer treatment agents linked to adverse CV effects include the antimetabolite 5-fluorouracil, which accelerates CAD; the selective proteasome inhibitor carfilzomib (Kyrprolis, Onyx), which has been associated with congestive HF, cardiac arrest and arrhythmias; the angiogenesis inhibitor bevacizumab (Avastin, Genentech), which, like some of the tyrosine kinase inhibitors, blocks vascular endothelial growth, increasing risk for hypertension and bleeding; and thalidomide (Thalomid, Celgene) and its derivative lenalidomide (Revlimid, Celgene), which are associated with increased risk for thromboembolism (see Table).
Radiation therapy “can cause accelerated CAD, valvular disease and pericardial disease, and these usually manifest themselves a decade after initial exposure,” Sahni said, noting that risks are particularly elevated in patients who receive both radiation and a chemotherapy drug known to have adverse CV effects.
It is important to continue researching how cancer treatments, especially new ones, might adversely affect the heart and the vascular system, Barac said.
“Any new drug that touches the pathways that are important for CV homeostasis is a potential CV hazard,” she said. “But we can’t blow this out of proportion. For example, the risk in young patients without risk factors if they are given anthracyclines is very small.”
The case for collaboration
Given the amount of evidence that is amassing, collaboration between cardiologists and oncologists to mitigate CV risks from cancer treatments is becoming necessary, experts said. Although some institutions have set up cardio-oncology programs to facilitate this collaboration, the issue has barely registered at others.
“Cardio-oncology is new and is not on the radar of many practicing oncologists around the country,” Holcombe said. “Oncologists understand that there are some cardiac ramifications of cancer and of cancer treatment, and they often refer patients to cardiologists, but are not really aware of the emerging specialty of cardio-oncology. A lot of places in [the United States] don’t have a cardiologist who specializes in that.”
According to Anita D. Szady, MD, director of cardio-oncology in the division of cardiovascular medicine at the University of Florida, Gainesville, “it is important that the next generation of cardiologists is trained to recognize the harmful cardiovascular effects of chemotherapeutic agents and how to best manage this growing patient population.”
To characterize the landscape and raise awareness, Barac and colleagues, on behalf of the American College of Cardiology, in June 2015 published an assessment of cardio-oncology clinical care delivery and education, including the results of a survey of chiefs of cardiology and program directors.
“The survey respondents recognized clinical relevance but emphasized lack of national guidelines, lack of funds and limited awareness and infrastructure as the main challenges for development and growth of cardio-oncology,” Barac and colleagues wrote.
Shortly before publication, the ACC created a member section in cardio-oncology, which the authors characterized as “a major step forward.”
“The best way to mitigate the CV risk is to get a cardiologist to understand the oncologic treatments that patients are receiving and get them involved early on so that you can get some baseline study and then compare later studies to what was obtained at baseline,” Holcombe said.
The process needs to begin with patient education about the potential cardiotoxicity of their cancer treatments and the signs and symptoms of heart disease, according to Sahni. “Once a patient reports to you that they are feeling different from these drugs, which could be attributed to heart disease, we can intervene and detect earlier,” Sahni said.
Oncology physicians and nurse practitioners should be trained on how to use tools like the cardiac risk score index and how to recognize which patients have abnormalities on ECG and echocardiography, so that they can quickly refer a patient to a cardiologist, she said.
“Patients undergoing potentially cardiotoxic chemotherapy should be monitored by incorporating techniques such as global longitudinal strain (GLS) in their surveillance echocardiograms, because the latest studies have suggested that a decline in GLS may predict cardiotoxicity even before the left ventricular ejection fraction declines,” Sahni said. Testing for certain biomarkers such as troponin I and myeloperoxidase could also be useful adjunctive tools to predict patients who are at higher risk for developing future cardiotoxicity, she added.
Once a patient’s CV risk is known, cardiologists and oncologists can weigh that against the cancer risks and “tailor-make the patient’s therapy to his or her CV risk profile,” Sahni said.
In addition to consultation, cardiologists and oncologists need to collaborate on research to help develop best practices for detection and treatment of cancer treatment-related CV conditions, Hundley said.
“Today, there are not many primary or secondary prevention algorithms for preventing CVD [after cancer treatment], although that looks like the primary cause of morbidity and mortality for these patients,” he said. “That’s why linking these two groups together to perform this research is important, so that we can design these preventive measures. The problem is, we don’t have the data or the know-how to implement guidelines because our management strategy is basically 30 years old.”
Importance of primary prevention
One of the next major steps in the evolution of cardio-oncology is determining whether patients can be started on medications before or during their cancer treatments that will protect their heart and vascular system against any toxic effects from the cancer treatments.
“Early detection is important, but prevention of cardiotoxicity is of event greater importance. Early involvement of the cardiologist is key to not missing this opportunity,” Szady said.
In the OVERCOME trial, published in 2013, Xavier Bosch, MD, PhD, from Thorax Institute, Hospital Clinic, Barcelona, Spain, and colleagues found that patients with acute leukemia or malignant hemopathies assigned enalapril and carvedilol had no change in LVEF, but those assigned no CV treatment experienced a decline in LVEF at 6 months (absolute difference on echocardiography, –3.1%; P = .035; absolute difference on cardiac MR, –3.4%; P = .09). In addition, compared with controls, at 6 months the intervention group had a lower rate of death or HF (6.7% vs. 22%; P = .036) and death, HF or final LVEF less than 45% (6.7% vs. 24.4%; P = .02).
Results of the PRADA trial, conducted in women with breast cancer, were presented at the American Heart Association Scientific Sessions in November 2015. Candesartan, an angiotensin receptor blocker, prevented LVEF decline, but the beta-blocker metoprolol was not protective in women treated for breast cancer with anthracycline or taxanes, trastuzumab if they were HER-2–positive, as well as radiation in some cases. During a follow-up of 10 to 61 weeks, LVEF declined 2.6% (95% CI, 1.5-3.8) in those assigned candesartan and 0.8% in those assigned placebo (95% CI, –0.4 to 1.9; P for between-group difference = .026). A similar difference was not observed between metoprolol and placebo.
The MANTICORE trial, presented at the 2015 San Antonio Breast Cancer Symposium, tested whether taking bisoprolol, a beta-blocker, or perindopril, an ACE inhibitor, for 1 year was cardioprotective in women assigned trastuzumab for HER-2–overexpressing early-stage breast cancer. At 2 years, compared with placebo, those assigned bisoprolol had a lower decline in LVEF (–1% vs. –4%; P < .05), although neither intervention had a significant effect on trastuzumab-induced LV remodeling, researchers found. In addition, the bisoprolol and perindopril groups each had only one patient interrupt trastuzumab due to LV dysfunction, compared with eight in the placebo group (P = .002).
“It’s not the same whether you start someone on a beta-blocker or an ACE inhibitor,” Barac said. “There is nothing set in stone. That is one reason why it is important to get the cardiologist involved early.”
Studies such as these are “the mainstay of emerging research in this field,” Sahni said, because “even before the advent of HF from cardiotoxic chemotherapy, we can actually mitigate the CV risk and reduce the development of HF with these drugs.”
Future heart issues
Another aspect of CV care in patients who have undergone cancer treatment is that the longer they survive, the more likely they may need cardiac intervention later in life. With that in mind, the Society for Cardiovascular Angiography and Interventions recently published an expert consensus statement on management and treatment of patients with cancer presenting in the cath lab with CVD.
“What [SCAI] realized was that in the community, there was an obstacle to tackling patients with both cancer and heart disease because of the complexity of them as well as factors such as anemia and thrombocytopenia,” statement author Cezar A. Iliescu, MD, FACC, FSCAI, director of the cardiac catheterization laboratory at The University of Texas MD Anderson Cancer Center, told Cardiology Today. “SCAI looked at centers and operators that had experience in doing these procedures, and saw how they did them safely. The beauty of the document is that it encourages interventionalists throughout the country to help patients that have cancer in the effort to overcome CV problems.”
The document lists 20 chemotherapeutic agents associated with myocardial ischemia and posits an algorithm for screening people for PCI based on prior chemotherapy and radiation exposure. It also shows that using fractional flow reserve to assess plaque vulnerability can defer intervention in more than 40% of patients with cancer, consistent with the results from the DEFER and FAME studies, and offers guidance on how to properly perform cardiac catheterization in patients with cancer, who because of thrombocytopenia are at increased risk for clotting and bleeding.
“We’re trying not to intervene and interfere with cancer therapy as much as we can,” Iliescu said. “But on the other hand, [the document] gives interventional cardiologists the tools in such a situation. ... [PCI] is a little more meticulous in such a situation because of the increased frailty of the patient population and the added comorbidities.”
Further research needed
The field of cardio-oncology is on its way to being established, and much more is known today than just 5 years ago, but numerous research gaps remain.
Following in the footsteps of the OVERCOME, PRADA and MANTICORE trials are more studies of cardioprotective therapies administered before or during cancer treatment. Although the aforementioned studies analyzed patients with normal heart function before cancer therapy, the SAFE-HEaRt study will evaluate whether treatment of breast cancer with HER-2–targeted therapies is feasible in patients with slightly reduced LVEF if they take beta-blockers or ACE inhibitors beforehand. “SAFE-HEaRt is a cardiac safety trial in patients who otherwise wouldn’t be eligible for their therapies because their EF is lower,” said Barac, an investigator for that study.
There also should be more assessment of CV outcomes in phase 3 trials of new cancer treatments because “we would have more understanding of why the drugs have certain effects,” Barac said. “By the time the FDA announces that there might be a signal, it may be very late and represent a lost opportunity.”
More studies that would help in the development of risk prediction are also necessary, she said. For example, it would be useful to be able to combine a risk-prediction tool for cardiomyopathy with risk-prediction tools used by oncologists, she said.
Sahni said she would like to see more study of the long-term effects of cardioprotective agents in this population, now that patients with cancer are surviving longer than before. In particular, she said, the field needs to know whether lessons learned from studies of childhood cancer survivors, such as dexrazoxane (Zinecard, Pfizer) reducing CV risks in survivors of childhood leukemia and lymphoma, can be applied to adult survivors.
Recognizing the importance of the issue, the NHLBI and the National Cancer Institute have issued a joint program announcement calling for research proposals for improving outcomes in cancer treatment-related cardiotoxicity.
Hundley said funds should be appropriated in three areas: basic science research to identify who is most at risk, population-based studies to identify risk factors and whether changing them because of cancer therapy boosts the development of CVD and other heart diseases, and clinical trials of therapeutic interventions that could prevent CV conditions and events after cancer treatment.
A changed dynamic
Much more research needs to be done, but enough has been learned that the dynamic between cardiologists and oncologists has forever changed.
“The threshold to get cardiology involved has decreased,” Iliescu said. “It used to be that patients were referred to cardiology when they were extremely symptomatic and when they were showing signs of having HF, MI or cardiac tamponade. The collaboration is moving from where we were treating people who were decompensating that, in many cases, the battle was lost, to where we are preventing, screening, detecting and facing these problems. In a way, it brings more complex decisions, but in the long run, it will definitely translate to better outcomes for the patients.”
Some institutions have adopted this attitude and created a cardio-oncology program or a structured relationship between cardiology and oncology. However, many others have not yet caught on. But the question remains: At what cost?
“You don’t want to trade cancer, which is a manageable disease, for another disorder, CVD,” Hundley said. “Working together will become the way to solve this in the future.” – by Erik Swain
- Armenian SH, et al. J Clin Oncol. 2016;doi:10.1200/JCO.2015.64.0409.
- Barac A, et al. J Am Coll Cardiol. 2015;doi:10.1016/j.jacc.2015.04.059.
- Bosch X, et al. J Am Coll Cardiol. 2013;doi:10.1016/j.jacc.2013.02.072.
- Douxfils J, et al. JAMA Oncol. 2016;doi:10.1001/jamaoncol.2015.5932.
- Gulati G, et al. Late-Breaking Clinical Trials 4. Presented at: American Heart Association Scientific Sessions; Nov. 7-11, 2015; Orlando, Fla.
- Gunn HM, et al. J Adolesc Young Adult Oncol. 2015;doi:10.1089/jayao.2015.0036.
- Iliescu CA, et al. Catheter Cardiovasc Interv. 2016;doi:10.1002/ccd.26379.
- Patnaik JL, et al. Breast Cancer Res. 2011;doi:10.1186/bcr2901.
- Pituskin E, et al. Abstract S1-05. Presented at: San Antonio Breast Cancer Symposium; Dec. 8-12, 2015; San Antonio.
- For more information:
- Ana Barac, MD, PhD, FACC, can be reached at MedStar Washington Hospital Center, 110 Irving St. NW, Suite 1F12 18, Washington, DC 20010; email: firstname.lastname@example.org.
- Randall Holcombe, MD, MBA, can be reached at Tisch Cancer Institute, Box 1128, 1 Gustave L. Levy Place, New York, NY 10029; email: email@example.com.
- W. Gregory Hundley, MD, FACC, FAHA, can be reached at Cardiovascular Medicine Division, Wake Forest Health Sciences, 1 Medical Center Blvd., Winston-Salem, NC 27157; email: firstname.lastname@example.org.
- Cezar A. Iliescu, MD, FACC, FSCAI, can be reached at 17400 Red Oak Drive, Houston, TX 77090; email: email@example.com.
- Gagan Sahni, MD, FACC, FACP, can be reached at The Mount Sinai Hospital, 1 Gustave L. Levy Place, Box 1030, New York, NY 10029; email: firstname.lastname@example.org.
- Anita D. Szady, MD, can be reached at 1600 SW Archer Road, Gainesville, FL 32608; email: email@example.com.
Disclosures: Barac, Holcombe, Hundley, Iliescu, Sahni and Szady report no relevant financial disclosures.