- Email: [email protected]
Cardio-oncology in clinical studies and real life
Seminars in Oncology 46 (2019) 421–425
Contents lists available at ScienceDirect
Seminars in Oncology journal homepage: www.elsevier.com/locate/seminoncol
Cardio-oncology in clinical studies and real life Susan F. Dent a, Thomas M. Suter b, Teresa López-Fernández c, Grzegorz Opolski d, Pierantonio Menna e, Giorgio Minotti e,∗ a
Duke University School of Medicine, Durham, North Carolina Swiss Cardiovascular Centre, Bern University, Bern, Switzerland Hospital Universitario La Paz, Instituto de Investigación La Paz-IdiPaz, Madrid, Spain d Medical University of Warsaw, Poland e University Campus Bio-Medico, Rome, Italy b c
a r t i c l e
i n f o
Article history: Received 6 January 2019 Accepted 9 January 2019
Keywords: Cardio-oncology Networks Clinical trials Real life Systolic dysfunction Diastolic dysfunction
a b s t r a c t Session V of the Colloquium was chaired by Professors Teresa López-Fernández of Spain and Grzegorz Opolski of Poland. The 3 speakers addressed cardio-oncology issues as they relate to both clinical studies and real life situations. Professor Susan Dent discussed cardio-oncology networks for patients, emphasizing the importance of establishing a framework where the expertise of the cardiology consultant can supplement and reinforce the goals of optimal cancer therapy. Professor Thomas Suter moved the discussion further, sharing his insight into cardiac monitoring in clinical trials, emphasizing the lack of uniform criteria and lack of consensus regarding reversibility of cardiac events and long-term implications of modest declines in systolic function frequently found in clinical trials for which long-term follow-up may not be a component of the trial. Professor Giorgio Minotti added important considerations to the discussion of clinical trials. He emphasized that the usual reporting of cardiac systolic function omits important diastolic dysfunction data generated but often ignored during the routine cardiac exams. The inclusion of cardiac biomarker changes would also help to broaden the perspective of cardiac effects and events seen in patients enrolled in clinical trials. © 2019 Elsevier Inc. All rights reserved.
Session summary Session V of the Second International Colloquium on CardioOncology, chaired by Drs. Lopez-Fernadez (Madrid, Spain) and Opolski (Warsaw, Poland), provided unique insights into cardiac aspects of supportive care for cancer patients. The subject of this session was not directly related to the evaluation of patients undergoing standard treatments for their malignancy, but focused on several aspects of management of the individual patient. The importance of accumulating and integrating data related to cardiac aspects of cancer care, the pitfalls as well as the benefits of surveillance, and how inclusion into clinical trials may improve the ultimate outcome for patients, and also enhance the patient experience was discussed. This one hopes will optimize the ultimate benefit of individual interventions as well as provide strategies to improve overall care for future generations of those afflicted with cancer.
∗ Corresponding author. University Campus Bio-Medico, Via Alvaro del Portillo 21, 00128 Rome, Italy, Tel.: 01139062215419109 E-mail address: [email protected] (G. Minotti).
https://doi.org/10.1053/j.seminoncol.2019.01.004 0093-7754/© 2019 Elsevier Inc. All rights reserved.
The session opened with Professor Susan F. Dent’s discussion of Cardio-Oncology Networks for Patients. With cancer and cardiovascular disease as the 2 leading causes of mortality in North America, and with advances in cancer treatment that have led to a significant increase in the number of cancer survivors, long-term toxicity from cancer therapy has emerged as a growing concern [1]. By 2026, it is estimated there will be over 25 million cancer survivors in the United States alone [2]. Strategies to identify patients at greatest risk of cancer therapy-related cardiac dysfunction are needed in order to minimize adverse cardiac events during and following cancer treatment. Over the past decade, cardiooncology has evolved as a subspecialty in medicine where health care providers, including oncologists, cardiologists, nurses, and allied health care professionals, work together to ensure optimal clinical outcomes for patients receiving cancer treatment. Dedicated cardio-oncology clinics have emerged across North America and internationally to manage patients who are at risk of developing, or have developed cardiotoxicity related to their cancer therapy (Fig. 1). The successful implementation of a multidisciplinary cardio-oncology clinic is complex. First, the need for a cardio-oncology clinic must be clearly outlined, the scope of prac-
422
S.F. Dent, T.M. Suter and T. López-Fernández et al. / Seminars in Oncology 46 (2019) 421–425
Fig. 1. A depiction of a number of components in a cardio-oncology program, showing the inter-relation and integration of such components rather than as identifying them as isolated entities (Courtesy of Professor S.F. Dent).
tice must be defined and a vision must be communicated to provide the necessary ongoing support. Secondly, institutional support is paramount in developing the infrastructure needed to move forward. Thirdly, recruitment of oncologists, hematologists, and cardiologists, with an interest and expertise in cardio-oncology is essential in establishing a successful cardio-oncology program as well as the support from allied health care professionals including nurses and pharmacists [3]. In a successful cardio-oncology program, team members work together to identify patient care needs with the goal of optimizing cancer therapy without compromising cardiovascular health [4]. As we learn more about cancer therapy-related cardiotoxicity, specific treatment plans can be implemented and tailored to meet the individual needs of patients. Clinical risk factors, cardiac biomarkers, and cardiac imaging either alone or in combination, can determine which patients are at the highest risk of developing cardiac dysfunction or injury. These high-risk patients can be referred for prompt cardiology evaluation in order to optimize risk factors and cardiac health prior to the initiation of cancer therapy. This ensures patient safety, gives the care team the ability to anticipate any adverse outcome, and facilitates patients completing a full course of potentially life-saving treatment that otherwise may have been stopped or modified due to cardiovascular complications [5]. In order to improve our understanding of the cardiovascular complication of cancer therapies it is vital that established and emerging multidisciplinary cardio-oncology clinics report outcome data including: (a) patient demographics; (b) referral patterns; (c) cardiac investigations; (d) cardiac treatments; (f) clinical outcomes; and (g) the patient experience. Reporting clinical outcomes will enable us to provide better multidisciplinary management of cancer patients while maintaining the upward trend of cancer survivorship without compromising cardiac health. Importantly, the establishment of quality indicators and patient reported outcomes for cardio-oncology clinics should provide important information on the success of these programs. While the clinical care and support of patients during, and following their cancer therapy remains our primary goal, education of patients and health care providers is essential in order to empower individuals and health care providers with the knowledge needed to deliver the best cancer therapies while optimizing cardiovascular health. In an online survey of 444 cardiologists, 39% said they did not feel confident in dealing with cardiovascular needs specific to cancer patients [6]. While in a recent international cardiooncology survey of 160 health care providers (cardiologists—53.8%;
oncologists—32.5%), over half (55.8%) of cardiologists felt that they should provide monitoring for cardiotoxicity in cancer patients, even in the absence of symptoms. In contrast, 50.0% of oncologists felt that cardiologists should be involved only when patients develop cardiotoxicity. The majority of cardiologists (88.3%) believed that prognosis for cancer patients would be significantly improved with a cardio-oncology service while only 45.8% of oncologists shared this opinion, highlighting the need for education in this new field of medicine [7]. The International Cardio-Oncology Society, was founded with the intention of bringing awareness and education to institutions on an international scale. The Global CardioOncology Summit, an annual international conference, fosters communication, and education of health care providers involved in the care of cancer patients receiving potentially cardiotoxic cancer therapy. Globally, countries have established cardio-oncology networks (eg, Canadian Cardiac Oncology Network, British CardioOncology Society) to foster collaboration and education. In addition to promoting awareness and education among health care providers, it is vitally important that cancer patients be engaged in the development of patient-directed educational tools in cardio-oncology. These tools will empower individuals to be active participants in their cancer and cardiovascular care. As modern targeted cancer therapeutic strategies continue to emerge translational and clinical research is essential to improve our understanding of the exact mechanisms of cardiotoxicity as well as the potential therapeutic approaches to prevent and treat cancer therapy-related cardiac dysfunction. Research is needed to determine how to: (a) best identify cancer patients at greatest risk of cardiotoxicity employing imaging strategies and biomarkers; (b) prevent cardiotoxicity through prevention studies in high risk patients; (c) treat those who have developed cardiotoxicity by providing optimal cardiac therapy so that these patients can complete their cancer therapy. Research will also allow the cardiooncology clinic to gain recognition as an important concept [8]. For both patients and providers, initiatives such as the SURVIVE (cardiovaScUlaR toxicity in cancer and improVement In recoVEry) registry will further advance our knowledge and understanding of how to provide optimal care for this challenging patient population. Session V proceeded with a presentation by Dr. Thomas M. Suter from The University of Bern, Switzerland, addressing the strengths and weaknesses of cardiac surveillance in clinical trials. It is recognized that patients in clinical trials constitute a diverse population even when inclusion and exclusion criteria are restrictive. In that regard variation among the research population must be anticipated, and the predictive value of any single piece of data may be subject to interpretation, imprecision, or false positive adjudication. And while in various clinical trials, data related to thousands of patients has been reported, controversy still exists as to the extent of cardiac injury, even in the case of a drug that has been studied in multiple trials and in a variety of combinations; this has been especially the case where cardiac events are rare or where they may be reversible [9]. Nevertheless, important information has been obtained from these trials, yet aspects of them have made comparisons between the various studies problematic, and, in some instances, the conclusions uncertain. Clinical trials have inherent strengths, and even when they lack consistency with regard to cardiac monitoring, they attempt to be well controlled, incorporate rigorous and consistent criteria for cardiac monitoring, and are often well-funded. When trials are designed to quantitate cardiac events, they meet the criteria for sample size to confidently answer the cardiac question, but when they are part of a trial to evaluate oncologic efficacy they may be statistically under-powered to answer relevant questions regarding cardiac adverse events [10]. Furthermore, differences amongst trials in design and in criteria for adjudicating an event make the anal-
Months
423
72
54
63
45
27
36
9
CTX
18
60 40 20 0
Pre
LVEF %
CTX
63 72
60 40 20 0
Pre 9 18 27 36 45 54
63
72
LVEF (%)
Months
54
36
45
27
9
CTX
18
60 40 20 0
Pre
LVEF (%)
S.F. Dent, T.M. Suter and T. López-Fernández et al. / Seminars in Oncology 46 (2019) 421–425
Months
Fig. 2. Three patterns of left ventricular ejection fraction (LVEF) response seen in clinical trials. Left: a dip and recovery in LVEF is shown; center: a dip with partial recovery followed by a gradual further decline of LVEF, and right: an initial stable LVEF follow by a decline. (Courtesy of Professor T.M. Suter) CTX, cardiotoxic cancer therapy.
Fig. 3. The various treatment schedules of the 5 major adjuvant trastuzumab breast cancer trials. Note that in the FinHER trial the anthracycline was administered after rather than before trastuzumab (Courtesy of Professor T.M. Suter).
ysis of multiple trials together problematic. Different cardiac abnormalities are present in enrolled patients, and findings beyond the recognition of a decreased left-ventricular ejection fraction are important and sometimes not appreciated. Among these other parameters are disturbances of rhythm, thrombo-embolic events and coagulation abnormalities, ischemia, pericardial inflammation, and infiltrative myocarditis. In considering contractile dysfunction, 3 variants that can only be appreciated over time should be considered: (i) a transient decline in ejection fraction with return to baseline that is then maintained over time; (ii) a decline with incomplete return to baseline, followed by a gradual relentless further decline in the ejection fraction; and (iii) no initial decline, but a late gradual but relentless decline (Fig. 2). While the first of these forms may have an excellent prognosis, it may be the pattern that is more frequently adjudicated as a cardiac event. The third type may only be recognized when long-term follow-up studies of cardiac function are included. In cases of late cardiac dysfunction, the question as to whether we are seeing (i) a late effect of an initial subclinical insult; (ii) an ongoing stress that ultimately exceeds the heart’s ability to compensate; or (iii) a cardiac insult totally unrelated to the cancer or
the treatment being evaluated in the clinical trial can be difficult to discern. The adjuvant trastuzumab breast cancer trials, some of which have now had long-term cardiac events reported, provide an example of some of the elements that may confuse interpretation (Fig. 3). Of especial interest is the fact that the NSABP B031 trial had a 6-fold higher incidence of cardiac events than were reported in the HERA trial. Further analysis, however, offers considerable insight regarding this difference: patients enrolled in the HERA trial were randomized after anthracycline exposure and required an entry ejection fraction of ≥55%, thereby selecting a population with fewer underlying cardiac risks, and the time period between the last anthracycline exposure and the start of trastuzumab was extended to 89 days to allow for intercurrent radiation. Interestingly, in the HERA trial, 80% of those who experienced a cardiac event had favorable outcomes. The adjuvant trials clearly demonstrate that we are not yet able to identify those who will experience events, or those for whom the events have important subsequent consequences. Additionally, difficulties in interpreting cardiac events in oncologic clinical trials go beyond primary declines in ejection fraction.
424
S.F. Dent, T.M. Suter and T. López-Fernández et al. / Seminars in Oncology 46 (2019) 421–425
Fig. 4. Early cardiotoxicity induced by chemotherapy in low risk patients. Panel A shows that 29 of 80 comorbid-free cancer patients presented with Nt-proBNP elevations, impaired myocardial relaxation (grade I diastolic dysfunction, DD) and/or cardicac troponin elevations (cTnI) at a cardiologic assessement done as early as 1 week after the the last cycle of standard dose anthracycline-based or nonanthracycline chemotherapy. Note that grade I DD and Nt-proBNP elevations were mutually exclusive. Panel B shows that all patients presented with a preserved >50% LVEF. T0, baseline; T1, 1 week after chemotherapy. Adapted from Calabrese et al [11]. Nt-proBNP, N-terminal-pro hormone B-type natriuretic peptide.
In a large retrospective review of cardiac events related to sunitinib, patients showed a high degree of reversibility of their cardiac dysfunction with adequate control of sunitinib-associated hypertension and systolic dysfunction. In order for the cardio-oncologist to derive maximal information regarding cardiac events and their implications, trials must be well controlled, include an adequate number of patients, with rigorously defined criteria for cardiac events, and with sufficient funds to allow for long-term follow-up of late cardiac events. When trials are geographically problematic, or do not have sufficient cardiac long-term surveillance, they are much more likely to result in weak or confusing data. Ideally, trials should have longterm cardiac surveillance, and not end prematurely when only the oncologic question has been adequately solved but without consideration of late cardiac sequelae. The latter is understandably difficult and costly. The final presentation, Clinical studies with a real life approach: An Italian experience by Professor Giorgio Minotto from Rome, Italy provided an important perspective regarding the need to include additional and readily available cardiac data in oncologic clinical trials. Information related to diastolic dysfunction is an example of data that is often underutilized, while elevations of N-terminal-pro hormone B-type natriuretic peptide and/or troponin I are examples of data that could be gathered and used more often (TEXT BOX 1). Diastolic dysfunction, as defined by the parameters E to A waves ratio and E wave deceleration time, can precede declines in ejection fraction (Fig. 4, TEXT BOX 2) [11]. The interesting biologic activity of B-type natriuretic peptide (BNP) as a cause of cyclic guanosine monophosphate (cGMP)-dependent cardiac relaxation and as a metabolically active molecule associated with clinically relevant tachycardia have been underscored [12]. BNP elevations can have positive effects on the rate of myocardial relaxation or lusitropy, as well as chronotropic effects leading to an increase in the heart rate, and these effects might be explained by activation of natriuretic receptor-associated guanylyl cyclase and production of cGMP in ventricular myocytes and sinoatrial node, respectively. This relationship, depicted in Fig. 5, describes a pharmacologic scenario in which cancer drugs can lead to early diastolic dysfunction that in some patients is both heralded and modulated by BNP elevations. Patients with demonstrable BNP elevations should be considered for adequate pharmacologic treatment to correct the diastolic dysfunction and tachycardia.
Fig. 5. Simplified pathway depicting the feedback mediated by BNP elevations on diastolic function and the relationship with increased cardiac rate (see the text). From Menna et al, reprinted with permission from American Society for Pharmacology and Experimental Therapeutics [12]. BNP, brain natriuretic peptide.
TEXT BOX 1. Brain natriuretic peptide (BNP) • • • •
Also known as B-type natriuretic peptide. 32-amino acid polypeptide. Widely used as a diagnostic marker of systolic dysfunction. Secreted by cardiac myocytes in response to stretching caused by increased ventricular volume and/or wall distress. • Secreted attached to a 76–amino acid N-terminal fragment in the NT-proBNP (BNPT) prohormone which is biologically inactive. • Once released, BNP binds to and activates NRPA, the atrial natriuretic factor, and to a lesser extent NRPB similar to atrial natriuretic peptide (ANP) but with 10-fold lower affinity. • The biological half-life of BNP is twice as long as that of ANP and that of NT-proBNP is even longer, making
S.F. Dent, T.M. Suter and T. López-Fernández et al. / Seminars in Oncology 46 (2019) 421–425
BNP and NT-proBNP better targets than ANP for diagnostic blood testing. • By decreasing blood pressure and in turn systemic vascular resistance and afterload, BNP leads to an increase in natriuresis. • However, both BNP and ANP can result in a decrease in cardiac output due to an overall decrease in central venous pressure and preload as a result of the reduction in blood volume that follows natriuresis and diuresis.
• •
• • •
The N-terminal prohormone of brain natriuretic peptide (NT-proBNP or BNPT) A non-active prohormone with an inactive 76 amino acid N-terminal portion that is cleaved from the molecule to release brain natriuretic peptide (BNP) Both BNP and NT-proBNP levels in the blood are used for screening, diagnosis of acute congestive heart failure and may be useful to establish prognosis in heart failure, as both markers are typically higher in patients with worse outcome Plasma concentrations of both BNP and NT-proBNP are also typically increased in patients with asymptomatic or symptomatic left ventricular dysfunction Both BNP and NT-proBNP are released in response to changes in pressure inside the heart. These changes can be related to heart failure and other cardiac problems. Both BNP and NT-proBNP are increased in patients with symptomatic or asymptomatic diastolic dysfunction
TEXT BOX 2. Diastolic dysfunction occurs as a result of both functional and structural causes. Diastole is affected by many factors. Before evaluating diastolic function one must differentiate between patients with normal left ventricular EF and reduced left ventricular EF or structural heart disease. In patients with reduced EF one must evaluate peak E wave velocity and E/A ratio. The E wave reflects the passive filling of the LV with the A wave reflecting the filling that occurs with active contraction of the left atrium in atrial systole. The normal pattern of diastolic filling is for the E-wave to be taller than the A-wave. In assessing diastolic dysfunction, the peaks of the E and A waves should be assessed, and an E/A ratio should be calculated. In patients with reduced LVEF: • E/A ratio ≤0.8 with peak E velocity of ≤50 cm/sec = normal filling pressures = Grade I diastolic dysfunction. • E/A ratio >2 = elevated left atrial filling pressures = Grade III diastolic dysfunction. • • E/A ratio ≤0.8 with a peak E velocity >50 cm/sec or E/A ratio >0.8 require additional evaluation including peak tricuspid regurgitation (TR) velocity, left atrial (LA) maximum volume index and E/e´ratio to assist in assigning the grade of dysfunction.
425
Additionally, BNP limits troponin release, a concept that adds fundamentally important insight into the understanding of troponin elevations in patients who are treated with anthracyclines. These concepts dispel the previously oversimplified concept of a simple relationship between anthracycline-associated myocyte apoptosis/necrosis and troponin release [13]. Declaration of Competing Interest No conflict of interest to declare References [1] Yeh ET, Bickford CL. Cardiovascular complications of cancer therapy. J Am Coll Cardiol 2009;53(24):2231–47. doi:10.1016/j.jacc.2009.02.050. [2] Miller KD, Siegel RL, Lin CC, et al. Cancer treatment and survivorship statistics, 2016. CA Cancer J Clin 2016;66(4):271–89. doi:10.3322/caac.21349. [3] Snipelisky D, Park JY, Lerman A, et al. How to develop a cardio-oncology clinic. Heart Fail Clin 2017;13(2):347–59. doi:10.1016/j.hfc.2016.12.011. [4] Parent S, Pituskin E, Paterson DI. The cardio-oncology program: a multidisciplinary approach to the care of cancer patients with cardiovascular disease. Can J Cardiol 2016;32(7):847–51. doi:10.1016/j.cjca.2016.04.014. [5] Dent S, Law A, Aseyev O, Ghosh N, Johnson C. Co-ordinating cardio-oncology care. Cardio-oncology: principles, prevention and management. Elsevier; 2017. p. 221–36. [6] Barac A, Murtagh G, Carver JR, et al. Cardiovascular health of patients with cancer and cancer survivors. A road map to the next level. J Am Coll Cardiol 2015;65(25):2739–46. doi:10.1016/j.jacc.2015.04.059. [7] Peng J, et al. An International Survey of Health Care Providers’ Knowledge of Cardiac Complications of Cancer Treatments. In: Accepted abstract for The Canadian Cardiovascular Conference, Toronto, Canada; October 20-23, 2018. [8] Okwuosa TM, Barac A. Burgeoning cardio-oncology programs. J Am Coll Cardiol 2015;66(10):1193–7. doi:10.1016/j.jacc.2015.07.033. [9] Suter TM, Ewer MS. Cancer drugs and the heart: importance and management. Eur Heart J 2013;34:1102–11. [10] Halpern SD, Karlawish JH, Berlin JA. The continuing unethical conduct of underpowered clinical trials. JAMA 2002;288:358–62. [11] Calabrese V, Menna P, Annibali O, et al. Early diastolic dysfunction after cancer chemotherapy: primary endpoint results of a multicenter cardio-oncology study. Chemotherapy 2018;63:55–63. [12] Menna P, Calabrese V, Armento G, et al. Pharmacology of cardio-oncology: chronotropic and lusitropic effects of B-type natriuretic peptide in cancer patients with early diastolic dysfunction induced by anthracycline or nonanthracycline chemotherapy. J Pharmacol Exp Ther 2018;366:158–68. [13] Menna P, Salvatorelli E, Armento G, et al. Minotti G The endogenous lusitropic and chronotropic agent, B-type natriuretic peptide, limits cardiac troponin release in cancer patients with an early impairment of myocardial relaxation induced by anthracyclines. J Pharmacol Exp Ther 2018;367:518–27.
Contents lists available at ScienceDirect
Seminars in Oncology journal homepage: www.elsevier.com/locate/seminoncol
Cardio-oncology in clinical studies and real life Susan F. Dent a, Thomas M. Suter b, Teresa López-Fernández c, Grzegorz Opolski d, Pierantonio Menna e, Giorgio Minotti e,∗ a
Duke University School of Medicine, Durham, North Carolina Swiss Cardiovascular Centre, Bern University, Bern, Switzerland Hospital Universitario La Paz, Instituto de Investigación La Paz-IdiPaz, Madrid, Spain d Medical University of Warsaw, Poland e University Campus Bio-Medico, Rome, Italy b c
a r t i c l e
i n f o
Article history: Received 6 January 2019 Accepted 9 January 2019
Keywords: Cardio-oncology Networks Clinical trials Real life Systolic dysfunction Diastolic dysfunction
a b s t r a c t Session V of the Colloquium was chaired by Professors Teresa López-Fernández of Spain and Grzegorz Opolski of Poland. The 3 speakers addressed cardio-oncology issues as they relate to both clinical studies and real life situations. Professor Susan Dent discussed cardio-oncology networks for patients, emphasizing the importance of establishing a framework where the expertise of the cardiology consultant can supplement and reinforce the goals of optimal cancer therapy. Professor Thomas Suter moved the discussion further, sharing his insight into cardiac monitoring in clinical trials, emphasizing the lack of uniform criteria and lack of consensus regarding reversibility of cardiac events and long-term implications of modest declines in systolic function frequently found in clinical trials for which long-term follow-up may not be a component of the trial. Professor Giorgio Minotti added important considerations to the discussion of clinical trials. He emphasized that the usual reporting of cardiac systolic function omits important diastolic dysfunction data generated but often ignored during the routine cardiac exams. The inclusion of cardiac biomarker changes would also help to broaden the perspective of cardiac effects and events seen in patients enrolled in clinical trials. © 2019 Elsevier Inc. All rights reserved.
Session summary Session V of the Second International Colloquium on CardioOncology, chaired by Drs. Lopez-Fernadez (Madrid, Spain) and Opolski (Warsaw, Poland), provided unique insights into cardiac aspects of supportive care for cancer patients. The subject of this session was not directly related to the evaluation of patients undergoing standard treatments for their malignancy, but focused on several aspects of management of the individual patient. The importance of accumulating and integrating data related to cardiac aspects of cancer care, the pitfalls as well as the benefits of surveillance, and how inclusion into clinical trials may improve the ultimate outcome for patients, and also enhance the patient experience was discussed. This one hopes will optimize the ultimate benefit of individual interventions as well as provide strategies to improve overall care for future generations of those afflicted with cancer.
∗ Corresponding author. University Campus Bio-Medico, Via Alvaro del Portillo 21, 00128 Rome, Italy, Tel.: 01139062215419109 E-mail address: [email protected] (G. Minotti).
https://doi.org/10.1053/j.seminoncol.2019.01.004 0093-7754/© 2019 Elsevier Inc. All rights reserved.
The session opened with Professor Susan F. Dent’s discussion of Cardio-Oncology Networks for Patients. With cancer and cardiovascular disease as the 2 leading causes of mortality in North America, and with advances in cancer treatment that have led to a significant increase in the number of cancer survivors, long-term toxicity from cancer therapy has emerged as a growing concern [1]. By 2026, it is estimated there will be over 25 million cancer survivors in the United States alone [2]. Strategies to identify patients at greatest risk of cancer therapy-related cardiac dysfunction are needed in order to minimize adverse cardiac events during and following cancer treatment. Over the past decade, cardiooncology has evolved as a subspecialty in medicine where health care providers, including oncologists, cardiologists, nurses, and allied health care professionals, work together to ensure optimal clinical outcomes for patients receiving cancer treatment. Dedicated cardio-oncology clinics have emerged across North America and internationally to manage patients who are at risk of developing, or have developed cardiotoxicity related to their cancer therapy (Fig. 1). The successful implementation of a multidisciplinary cardio-oncology clinic is complex. First, the need for a cardio-oncology clinic must be clearly outlined, the scope of prac-
422
S.F. Dent, T.M. Suter and T. López-Fernández et al. / Seminars in Oncology 46 (2019) 421–425
Fig. 1. A depiction of a number of components in a cardio-oncology program, showing the inter-relation and integration of such components rather than as identifying them as isolated entities (Courtesy of Professor S.F. Dent).
tice must be defined and a vision must be communicated to provide the necessary ongoing support. Secondly, institutional support is paramount in developing the infrastructure needed to move forward. Thirdly, recruitment of oncologists, hematologists, and cardiologists, with an interest and expertise in cardio-oncology is essential in establishing a successful cardio-oncology program as well as the support from allied health care professionals including nurses and pharmacists [3]. In a successful cardio-oncology program, team members work together to identify patient care needs with the goal of optimizing cancer therapy without compromising cardiovascular health [4]. As we learn more about cancer therapy-related cardiotoxicity, specific treatment plans can be implemented and tailored to meet the individual needs of patients. Clinical risk factors, cardiac biomarkers, and cardiac imaging either alone or in combination, can determine which patients are at the highest risk of developing cardiac dysfunction or injury. These high-risk patients can be referred for prompt cardiology evaluation in order to optimize risk factors and cardiac health prior to the initiation of cancer therapy. This ensures patient safety, gives the care team the ability to anticipate any adverse outcome, and facilitates patients completing a full course of potentially life-saving treatment that otherwise may have been stopped or modified due to cardiovascular complications [5]. In order to improve our understanding of the cardiovascular complication of cancer therapies it is vital that established and emerging multidisciplinary cardio-oncology clinics report outcome data including: (a) patient demographics; (b) referral patterns; (c) cardiac investigations; (d) cardiac treatments; (f) clinical outcomes; and (g) the patient experience. Reporting clinical outcomes will enable us to provide better multidisciplinary management of cancer patients while maintaining the upward trend of cancer survivorship without compromising cardiac health. Importantly, the establishment of quality indicators and patient reported outcomes for cardio-oncology clinics should provide important information on the success of these programs. While the clinical care and support of patients during, and following their cancer therapy remains our primary goal, education of patients and health care providers is essential in order to empower individuals and health care providers with the knowledge needed to deliver the best cancer therapies while optimizing cardiovascular health. In an online survey of 444 cardiologists, 39% said they did not feel confident in dealing with cardiovascular needs specific to cancer patients [6]. While in a recent international cardiooncology survey of 160 health care providers (cardiologists—53.8%;
oncologists—32.5%), over half (55.8%) of cardiologists felt that they should provide monitoring for cardiotoxicity in cancer patients, even in the absence of symptoms. In contrast, 50.0% of oncologists felt that cardiologists should be involved only when patients develop cardiotoxicity. The majority of cardiologists (88.3%) believed that prognosis for cancer patients would be significantly improved with a cardio-oncology service while only 45.8% of oncologists shared this opinion, highlighting the need for education in this new field of medicine [7]. The International Cardio-Oncology Society, was founded with the intention of bringing awareness and education to institutions on an international scale. The Global CardioOncology Summit, an annual international conference, fosters communication, and education of health care providers involved in the care of cancer patients receiving potentially cardiotoxic cancer therapy. Globally, countries have established cardio-oncology networks (eg, Canadian Cardiac Oncology Network, British CardioOncology Society) to foster collaboration and education. In addition to promoting awareness and education among health care providers, it is vitally important that cancer patients be engaged in the development of patient-directed educational tools in cardio-oncology. These tools will empower individuals to be active participants in their cancer and cardiovascular care. As modern targeted cancer therapeutic strategies continue to emerge translational and clinical research is essential to improve our understanding of the exact mechanisms of cardiotoxicity as well as the potential therapeutic approaches to prevent and treat cancer therapy-related cardiac dysfunction. Research is needed to determine how to: (a) best identify cancer patients at greatest risk of cardiotoxicity employing imaging strategies and biomarkers; (b) prevent cardiotoxicity through prevention studies in high risk patients; (c) treat those who have developed cardiotoxicity by providing optimal cardiac therapy so that these patients can complete their cancer therapy. Research will also allow the cardiooncology clinic to gain recognition as an important concept [8]. For both patients and providers, initiatives such as the SURVIVE (cardiovaScUlaR toxicity in cancer and improVement In recoVEry) registry will further advance our knowledge and understanding of how to provide optimal care for this challenging patient population. Session V proceeded with a presentation by Dr. Thomas M. Suter from The University of Bern, Switzerland, addressing the strengths and weaknesses of cardiac surveillance in clinical trials. It is recognized that patients in clinical trials constitute a diverse population even when inclusion and exclusion criteria are restrictive. In that regard variation among the research population must be anticipated, and the predictive value of any single piece of data may be subject to interpretation, imprecision, or false positive adjudication. And while in various clinical trials, data related to thousands of patients has been reported, controversy still exists as to the extent of cardiac injury, even in the case of a drug that has been studied in multiple trials and in a variety of combinations; this has been especially the case where cardiac events are rare or where they may be reversible [9]. Nevertheless, important information has been obtained from these trials, yet aspects of them have made comparisons between the various studies problematic, and, in some instances, the conclusions uncertain. Clinical trials have inherent strengths, and even when they lack consistency with regard to cardiac monitoring, they attempt to be well controlled, incorporate rigorous and consistent criteria for cardiac monitoring, and are often well-funded. When trials are designed to quantitate cardiac events, they meet the criteria for sample size to confidently answer the cardiac question, but when they are part of a trial to evaluate oncologic efficacy they may be statistically under-powered to answer relevant questions regarding cardiac adverse events [10]. Furthermore, differences amongst trials in design and in criteria for adjudicating an event make the anal-
Months
423
72
54
63
45
27
36
9
CTX
18
60 40 20 0
Pre
LVEF %
CTX
63 72
60 40 20 0
Pre 9 18 27 36 45 54
63
72
LVEF (%)
Months
54
36
45
27
9
CTX
18
60 40 20 0
Pre
LVEF (%)
S.F. Dent, T.M. Suter and T. López-Fernández et al. / Seminars in Oncology 46 (2019) 421–425
Months
Fig. 2. Three patterns of left ventricular ejection fraction (LVEF) response seen in clinical trials. Left: a dip and recovery in LVEF is shown; center: a dip with partial recovery followed by a gradual further decline of LVEF, and right: an initial stable LVEF follow by a decline. (Courtesy of Professor T.M. Suter) CTX, cardiotoxic cancer therapy.
Fig. 3. The various treatment schedules of the 5 major adjuvant trastuzumab breast cancer trials. Note that in the FinHER trial the anthracycline was administered after rather than before trastuzumab (Courtesy of Professor T.M. Suter).
ysis of multiple trials together problematic. Different cardiac abnormalities are present in enrolled patients, and findings beyond the recognition of a decreased left-ventricular ejection fraction are important and sometimes not appreciated. Among these other parameters are disturbances of rhythm, thrombo-embolic events and coagulation abnormalities, ischemia, pericardial inflammation, and infiltrative myocarditis. In considering contractile dysfunction, 3 variants that can only be appreciated over time should be considered: (i) a transient decline in ejection fraction with return to baseline that is then maintained over time; (ii) a decline with incomplete return to baseline, followed by a gradual relentless further decline in the ejection fraction; and (iii) no initial decline, but a late gradual but relentless decline (Fig. 2). While the first of these forms may have an excellent prognosis, it may be the pattern that is more frequently adjudicated as a cardiac event. The third type may only be recognized when long-term follow-up studies of cardiac function are included. In cases of late cardiac dysfunction, the question as to whether we are seeing (i) a late effect of an initial subclinical insult; (ii) an ongoing stress that ultimately exceeds the heart’s ability to compensate; or (iii) a cardiac insult totally unrelated to the cancer or
the treatment being evaluated in the clinical trial can be difficult to discern. The adjuvant trastuzumab breast cancer trials, some of which have now had long-term cardiac events reported, provide an example of some of the elements that may confuse interpretation (Fig. 3). Of especial interest is the fact that the NSABP B031 trial had a 6-fold higher incidence of cardiac events than were reported in the HERA trial. Further analysis, however, offers considerable insight regarding this difference: patients enrolled in the HERA trial were randomized after anthracycline exposure and required an entry ejection fraction of ≥55%, thereby selecting a population with fewer underlying cardiac risks, and the time period between the last anthracycline exposure and the start of trastuzumab was extended to 89 days to allow for intercurrent radiation. Interestingly, in the HERA trial, 80% of those who experienced a cardiac event had favorable outcomes. The adjuvant trials clearly demonstrate that we are not yet able to identify those who will experience events, or those for whom the events have important subsequent consequences. Additionally, difficulties in interpreting cardiac events in oncologic clinical trials go beyond primary declines in ejection fraction.
424
S.F. Dent, T.M. Suter and T. López-Fernández et al. / Seminars in Oncology 46 (2019) 421–425
Fig. 4. Early cardiotoxicity induced by chemotherapy in low risk patients. Panel A shows that 29 of 80 comorbid-free cancer patients presented with Nt-proBNP elevations, impaired myocardial relaxation (grade I diastolic dysfunction, DD) and/or cardicac troponin elevations (cTnI) at a cardiologic assessement done as early as 1 week after the the last cycle of standard dose anthracycline-based or nonanthracycline chemotherapy. Note that grade I DD and Nt-proBNP elevations were mutually exclusive. Panel B shows that all patients presented with a preserved >50% LVEF. T0, baseline; T1, 1 week after chemotherapy. Adapted from Calabrese et al [11]. Nt-proBNP, N-terminal-pro hormone B-type natriuretic peptide.
In a large retrospective review of cardiac events related to sunitinib, patients showed a high degree of reversibility of their cardiac dysfunction with adequate control of sunitinib-associated hypertension and systolic dysfunction. In order for the cardio-oncologist to derive maximal information regarding cardiac events and their implications, trials must be well controlled, include an adequate number of patients, with rigorously defined criteria for cardiac events, and with sufficient funds to allow for long-term follow-up of late cardiac events. When trials are geographically problematic, or do not have sufficient cardiac long-term surveillance, they are much more likely to result in weak or confusing data. Ideally, trials should have longterm cardiac surveillance, and not end prematurely when only the oncologic question has been adequately solved but without consideration of late cardiac sequelae. The latter is understandably difficult and costly. The final presentation, Clinical studies with a real life approach: An Italian experience by Professor Giorgio Minotto from Rome, Italy provided an important perspective regarding the need to include additional and readily available cardiac data in oncologic clinical trials. Information related to diastolic dysfunction is an example of data that is often underutilized, while elevations of N-terminal-pro hormone B-type natriuretic peptide and/or troponin I are examples of data that could be gathered and used more often (TEXT BOX 1). Diastolic dysfunction, as defined by the parameters E to A waves ratio and E wave deceleration time, can precede declines in ejection fraction (Fig. 4, TEXT BOX 2) [11]. The interesting biologic activity of B-type natriuretic peptide (BNP) as a cause of cyclic guanosine monophosphate (cGMP)-dependent cardiac relaxation and as a metabolically active molecule associated with clinically relevant tachycardia have been underscored [12]. BNP elevations can have positive effects on the rate of myocardial relaxation or lusitropy, as well as chronotropic effects leading to an increase in the heart rate, and these effects might be explained by activation of natriuretic receptor-associated guanylyl cyclase and production of cGMP in ventricular myocytes and sinoatrial node, respectively. This relationship, depicted in Fig. 5, describes a pharmacologic scenario in which cancer drugs can lead to early diastolic dysfunction that in some patients is both heralded and modulated by BNP elevations. Patients with demonstrable BNP elevations should be considered for adequate pharmacologic treatment to correct the diastolic dysfunction and tachycardia.
Fig. 5. Simplified pathway depicting the feedback mediated by BNP elevations on diastolic function and the relationship with increased cardiac rate (see the text). From Menna et al, reprinted with permission from American Society for Pharmacology and Experimental Therapeutics [12]. BNP, brain natriuretic peptide.
TEXT BOX 1. Brain natriuretic peptide (BNP) • • • •
Also known as B-type natriuretic peptide. 32-amino acid polypeptide. Widely used as a diagnostic marker of systolic dysfunction. Secreted by cardiac myocytes in response to stretching caused by increased ventricular volume and/or wall distress. • Secreted attached to a 76–amino acid N-terminal fragment in the NT-proBNP (BNPT) prohormone which is biologically inactive. • Once released, BNP binds to and activates NRPA, the atrial natriuretic factor, and to a lesser extent NRPB similar to atrial natriuretic peptide (ANP) but with 10-fold lower affinity. • The biological half-life of BNP is twice as long as that of ANP and that of NT-proBNP is even longer, making
S.F. Dent, T.M. Suter and T. López-Fernández et al. / Seminars in Oncology 46 (2019) 421–425
BNP and NT-proBNP better targets than ANP for diagnostic blood testing. • By decreasing blood pressure and in turn systemic vascular resistance and afterload, BNP leads to an increase in natriuresis. • However, both BNP and ANP can result in a decrease in cardiac output due to an overall decrease in central venous pressure and preload as a result of the reduction in blood volume that follows natriuresis and diuresis.
• •
• • •
The N-terminal prohormone of brain natriuretic peptide (NT-proBNP or BNPT) A non-active prohormone with an inactive 76 amino acid N-terminal portion that is cleaved from the molecule to release brain natriuretic peptide (BNP) Both BNP and NT-proBNP levels in the blood are used for screening, diagnosis of acute congestive heart failure and may be useful to establish prognosis in heart failure, as both markers are typically higher in patients with worse outcome Plasma concentrations of both BNP and NT-proBNP are also typically increased in patients with asymptomatic or symptomatic left ventricular dysfunction Both BNP and NT-proBNP are released in response to changes in pressure inside the heart. These changes can be related to heart failure and other cardiac problems. Both BNP and NT-proBNP are increased in patients with symptomatic or asymptomatic diastolic dysfunction
TEXT BOX 2. Diastolic dysfunction occurs as a result of both functional and structural causes. Diastole is affected by many factors. Before evaluating diastolic function one must differentiate between patients with normal left ventricular EF and reduced left ventricular EF or structural heart disease. In patients with reduced EF one must evaluate peak E wave velocity and E/A ratio. The E wave reflects the passive filling of the LV with the A wave reflecting the filling that occurs with active contraction of the left atrium in atrial systole. The normal pattern of diastolic filling is for the E-wave to be taller than the A-wave. In assessing diastolic dysfunction, the peaks of the E and A waves should be assessed, and an E/A ratio should be calculated. In patients with reduced LVEF: • E/A ratio ≤0.8 with peak E velocity of ≤50 cm/sec = normal filling pressures = Grade I diastolic dysfunction. • E/A ratio >2 = elevated left atrial filling pressures = Grade III diastolic dysfunction. • • E/A ratio ≤0.8 with a peak E velocity >50 cm/sec or E/A ratio >0.8 require additional evaluation including peak tricuspid regurgitation (TR) velocity, left atrial (LA) maximum volume index and E/e´ratio to assist in assigning the grade of dysfunction.
425
Additionally, BNP limits troponin release, a concept that adds fundamentally important insight into the understanding of troponin elevations in patients who are treated with anthracyclines. These concepts dispel the previously oversimplified concept of a simple relationship between anthracycline-associated myocyte apoptosis/necrosis and troponin release [13]. Declaration of Competing Interest No conflict of interest to declare References [1] Yeh ET, Bickford CL. Cardiovascular complications of cancer therapy. J Am Coll Cardiol 2009;53(24):2231–47. doi:10.1016/j.jacc.2009.02.050. [2] Miller KD, Siegel RL, Lin CC, et al. Cancer treatment and survivorship statistics, 2016. CA Cancer J Clin 2016;66(4):271–89. doi:10.3322/caac.21349. [3] Snipelisky D, Park JY, Lerman A, et al. How to develop a cardio-oncology clinic. Heart Fail Clin 2017;13(2):347–59. doi:10.1016/j.hfc.2016.12.011. [4] Parent S, Pituskin E, Paterson DI. The cardio-oncology program: a multidisciplinary approach to the care of cancer patients with cardiovascular disease. Can J Cardiol 2016;32(7):847–51. doi:10.1016/j.cjca.2016.04.014. [5] Dent S, Law A, Aseyev O, Ghosh N, Johnson C. Co-ordinating cardio-oncology care. Cardio-oncology: principles, prevention and management. Elsevier; 2017. p. 221–36. [6] Barac A, Murtagh G, Carver JR, et al. Cardiovascular health of patients with cancer and cancer survivors. A road map to the next level. J Am Coll Cardiol 2015;65(25):2739–46. doi:10.1016/j.jacc.2015.04.059. [7] Peng J, et al. An International Survey of Health Care Providers’ Knowledge of Cardiac Complications of Cancer Treatments. In: Accepted abstract for The Canadian Cardiovascular Conference, Toronto, Canada; October 20-23, 2018. [8] Okwuosa TM, Barac A. Burgeoning cardio-oncology programs. J Am Coll Cardiol 2015;66(10):1193–7. doi:10.1016/j.jacc.2015.07.033. [9] Suter TM, Ewer MS. Cancer drugs and the heart: importance and management. Eur Heart J 2013;34:1102–11. [10] Halpern SD, Karlawish JH, Berlin JA. The continuing unethical conduct of underpowered clinical trials. JAMA 2002;288:358–62. [11] Calabrese V, Menna P, Annibali O, et al. Early diastolic dysfunction after cancer chemotherapy: primary endpoint results of a multicenter cardio-oncology study. Chemotherapy 2018;63:55–63. [12] Menna P, Calabrese V, Armento G, et al. Pharmacology of cardio-oncology: chronotropic and lusitropic effects of B-type natriuretic peptide in cancer patients with early diastolic dysfunction induced by anthracycline or nonanthracycline chemotherapy. J Pharmacol Exp Ther 2018;366:158–68. [13] Menna P, Salvatorelli E, Armento G, et al. Minotti G The endogenous lusitropic and chronotropic agent, B-type natriuretic peptide, limits cardiac troponin release in cancer patients with an early impairment of myocardial relaxation induced by anthracyclines. J Pharmacol Exp Ther 2018;367:518–27.