Embed Size (px)
Citation preview
Activity and Outcomes of a Cardio-Oncology Service in the United Kingdom – a Five
Year Experience
Nilesh Pareek* MA MBBS MRCP1, Joaquim Cevallos* MD1, Pedro Moliner MD2, Mit Shah
Bsc MBBS1, Li Ling Tan1,3, Vicki Chambers Bsc RGN1, A. John Baksi PhD MRCP1, Rajdeep
S. Khattar MD FRCP1, Rakesh Sharma Bsc PhD FRCP1, Stuart D. Rosen MA MD FRCP1,4,
Alexander R. Lyon MA BM BCh PhD FRCP1,4.
1. Royal Brompton and Harefield NHS Foundation Trust, London, U.K.
2. Hospital Universitari Germans Trias i Pujol, Barcelona, Spain
3. Department of Cardiology, National University Heart Centre Singapore, Singapore.
4. National Heart and Lung Institute, Imperial College, London, U.K.
Institution where work performed: Royal Brompton and Harefield NHS Foundation Trust, London, U.K
Word Index: Cardiotoxicity, Cancer, Heart Failure,
Correspondence:
Alexander Lyon
Cardio-Oncology Service
Royal Brompton Hospital
Sydney Street,
London SW3 6NP
Tel: +44 (0)20 7352 8121
e-mail: [email protected]
First author: Nilesh Pareek and Joaquim Cevallos contributed equally to the manuscript
Short Title: Cardio-Oncology Service Development and Outcomes
Figures: 1
Tables: 5
Word Count: 3318 words
Abstract
1
1
2
3
4
5
6
7
8
9
10
1112
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
Aims
Cardio-Oncology clinics optimise the cardiovascular status of cancer patients but there is a
limited description of their structure, case mix, activity and results. The purpose of this paper
is to describe the activity and outcomes of a Cardio-Oncology service, particularly with
respect to supporting optimal cancer treatment and survival.
Methods and Results
We prospectively studied patients referred to our service from February 2011 to February
2016. NYHA class and parameters of cardiac function were measured at baseline and after
optimisation by our service. Up-titration of cardiac treatment, continuation of cancer therapy
and mortality were used as outcome measures.
Of the 535 patients (55.4% females) referred, rates of cardiotoxicity for anthracyclines, anti-
HER2 agents and tyrosine kinase inhibitors were 78.0%, 73.0% and 65.2% respectively.
Patients with left ventricular systolic dysfunction (LVSD) (n=128) were younger, had higher
rates of hypertension and previous exposure to chemotherapy/radiotherapy (p<0.001). At a
median follow-up of 360 days, 93.8% of the patients with LVSD showed improvement in
LVEF (45% Pre versus 53% Post (p <0.001)) and NYHA class (NYHA 3-4 in 22% Pre versus
10% Post (p = 0.01)). Eighty-nine percent of patients with LVSD were deemed fit for
continuation of cancer therapy after initiation of cardiac treatment. After adjusting for LVEF
and age, cardiovascular fitness and continuation of cancer treatment was associated with
lower mortality (HR-0.76; 95% CI-0.014–0.425;P<0.003)
Conclusions
Through the establishment of a Cardio-Oncology service, it is feasible to achieve high rates
of cardiac optimisation and cancer treatment continuation.
Key Questions
What is already known about this subject?
2
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
An increased prevalence of cardiovascular risk factors combined with improved cancer
survivorship and advances in cancer therapies has resulted in rising rates of myocardial
toxicity to cancer treatments with an increased demand for Cardio-Oncology services. A
small number of studies to date have shown that heart failure treatments can lead to
improvements in the cardiovascular status of patients with myocardial toxicity but the effects
on cancer treatment rates and survival outside research trials remain unknown.
What does this study add?
This study suggests that cardio-oncology services with standardised protocols of care can
successfully optimise the cardiovascular status of patients at high risk or with established
myocardial toxicity, resulting in high completion rates of cancer therapy and potentially
reduced mortality.
How might this impact on clinical practice?
This standardised model of care lends support to the concept of dedicated cardio-oncology
services for patients at high risk or with established myocardial toxicity but further studies
comparing outcomes against standard care are required.
3
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
INTRODUCTION
Cancer and cardiovascular (CV) disease are the most common causes of death in Europe
(1). Increasing numbers of cancer patients are surviving free of recurrence or living with the
disease owing to major advances in oncology treatments; an estimated 2.5 million people
are living with cancer in the UK (2). Improvements in cardiac and cancer treatments
combined with an ageing population have resulted in cancer being diagnosed in an older
cohort of patients with a higher burden of pre-existent CV disease (3). Several cancer
treatments, including chemotherapy agents, molecular targeted cancer therapies and
radiotherapy are effective but may cause significant cardiotoxicity (4). Cancer patients
therefore have a significantly increased risk of CV morbidity and mortality (5, 6).
An increased appreciation of the interaction between CV disease and cancer treatments has
led to the development of a new medical subspecialty - Cardio-Oncology (7). Cardio-
Oncology clinics have been developed with the purposes of 1) identifying patients at an
increased baseline CV risk before commencing cancer treatment; 2) minimising their risk
using primary prevention strategies; 3) diagnosis and treatment of myocardial toxicity during
or after cancer treatment; 4) supporting the patient through their cancer treatment regime
with a personalised surveillance programme; and 5) appropriate medium and long term
follow up for patients with CV disease attributed to cancer therapies.
As Cardio-Oncology services are relatively new, there has been limited description of their
activity and results. Here we report the five year consecutive experience of a Cardio-
Oncology service in the United Kingdom. We hypothesised that a personalised approach to
evaluation and optimisation of patients at high baseline risk or with established left
ventricular systolic dysfunction (LVSD) would limit cancer treatment interruption and improve
outcome.
METHODS
4
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
The Cardio-Oncology clinic is based at the Royal Brompton Hospital, London, United
Kingdom. The service is staffed by 3 Consultant Cardiologists (ARL, SR and RKS), a senior
clinical fellow in Cardio-Oncology (NP and JC) and a senior clinical nurse specialist (Band 8)
(VC), supported by specialist cardiac imaging with advanced echocardiography (RK) and
cardiovascular magnetic resonance (CMR) (JB). The service operates an initial one-stop
daycase service model in order to expedite assessment of cardiovascular status and provide
a rapid opinion to referrers. After establishment, the service was publicised by creation of a
service website (http://www.rbht.nhs.uk/patients/condition/cardio-oncology), dissemination of
referral documents to local centres and presentations at referring oncology hospitals,
oncology meetings and at Cardio-Oncology Conferences. Table 1 indicates common
reasons for referral to the service and our approach at baseline and follow-up. We
implemented a protocol where all new referrals attend for a baseline clinical assessment.
The latter incorporates blood tests including cardiac biomarkers, resting 12 lead
electrocardiogram (ECG) and a resting 2D trans-thoracic echocardiography (TTE) based on
British Society of Echocardiography recommendations in all patients. Stress
echocardiography (to assess for ischaemia and/or contractile reserve) and Cardiac Magnetic
Resonance (CMR) were performed in selected patients. CMR protocols applied to measure
LV and RV function, plus T2STIR and Late Gadolinium Enhancement (LGE) in order to
detect additional markers of clinical toxicity (ie fibrosis) or to assess cardiac masses and
their perfusion. TTE was used for serial comparison at follow-up using the same
echocardiography model and vendor.
We defined left ventricular dysfunction using our new Royal Brompton Hospital Myocardial
Toxicity definition (Table 2). This was developed as a practical clinical tool to incorporate
early biochemical and functional changes and to define a severity gradient from least severe
(Group 1) to the most severe (Group 6). Our classification is adapted from the Cardiac
Review and Evaluation Committee Reviewing Trastuzumab (CREC) criteria for myocardial
toxicity but includes more clinically relevant variables7. Elevation of biomarker levels was
defined as any value above the laboratory’s normal reference range (BNP > 20 ng/L and
5
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Troponin I > 20 ng/L). BNP was measured during follow-up of patients deemed at highest
risk of deterioration based on clinical opinion.
After multi-disciplinary team discussion and consultation, patients requiring current cancer
treatment were classified as either a) fit; b) at too high risk; or c) requiring further
cardiovascular optimisation to continue their cancer therapy, with recommendation for close
surveillance in higher risk patients from baseline or those with established cardiotoxicity if
ongoing cancer treatment was required (Table 2). Follow-up consultations with TTE to
assess LVEF and GLS (where indicated) were personalized for each patient depending on
current oncology treatment, LVEF, symptoms, cancer stage and response. In general, first
follow-up consultations were offered at 3 – 6 months following completion of cancer
treatment for low risk patients, at 2-4 weeks for those at high baseline risk or where oncology
treatment is ongoing but interrupted for cardiotoxicity to enable initiation and rapid up-
titration of guideline based heart failure treatment. The monitoring of higher risk patients
occurred at every 1-3 treatment cycles depending upon the severity of cardiotoxicity,
response to cardiac treatment and the dose and duration of planned further oncology
treatment. When appropriate, patients were prescribed statins and antiplatelet agents,
recommended moderate physical activity and lifestyle changes, as well as reviewed by our
senior clinical nurse specialist.
We specifically analysed 3 subgroups of patients: patients referred to our service due to an
exceedingly high CV risk; patients that developed any myocardial toxicity as per our
definitions (Groups 1-6) and those with significant LVEF reduction (Groups 5 and 6).
The primary outcomes measures were rates of cancer treatment at follow-up, as well as the
overall and cardiovascular mortality. Since many patients were expected deaths due to
cancer progression without post-mortems, we defined the cause of mortality as the most
likely cause as per patient clinical status at the last consultation. CV death was classified as
worsening CHF at last consultation and cancer related death where progressive metastatic
disease was present. When patients’ condition was stable from both a CV and oncological
perspective, the cause of death was labelled as unexpected (Table 5). The secondary CV
6
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
outcome measures in those with LVSD were a change in NYHA status and left ventricular
ejection fraction (LVEF) at follow-up, as well as successful up-titration of guideline based
congestive heart failure (CHF) treatments, defined as a numerical increase in equivalent
guideline based therapies for LVSD.
Data were collected prospectively and analysed with the SPSS statistical package. Normally
distributed data are represented as mean +/- standard deviation. Non-normally distributed
data are presented as medians with ranges or interquartile range (IQR). Comparisons were
performed with Student's T test for normally distributed variables; Chi-square for proportions
and Mood’s median test for non-normally distributed variables. Multi-variate logistic
regression analysis is presented as hazard ratios (HRs) with 95% confidence intervals (CIs).
A p < 0.05 was considered statistically significant.
RESULTS
Patient Baseline Characteristics
Five hundred and thirty-five patients (55.5% females) were referred between 1st February
2011 and 1st February 2016. Baseline characteristics of the patients are listed in Table 3.
Pre-existing CV risk factors were common, reflecting referral patterns to the service. Patients
with 31 types of cancer were referred; 171 patients (31.9%) had metastatic cancer
(Supplementary Figures 1a and 1b). Eighteen patients (3.4%) had evidence of primary or
metastatic cardiac involvement, the majority of which were sarcomas. Fifty-four patients
(10.1%) had a second malignancy; of this cohort 38 patients (70.4%) had received prior
chemotherapy.
We identified 128 patients (23.9%) with evidence of LVSD as per our classification (Groups
5 and 6). Patients in this group were more likely to be younger, with a lower burden of CV
risk factors and significantly higher previous exposure to chemotherapy and radiotherapy
(p<0.001). Amongst routine cardiac biomarkers, the frequency of elevated serum Brain
7
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
Natriuretic Peptide (BNP) levels at referral did not differ between groups but there was a
statistical difference in the rates of elevated serum cardiac troponin I.
Cardiac structural abnormalities were seen in 22% of patients with normal LVEF – including
valvular disease, pericardial effusions and/or cardiac masses. LGE was detected in 100 of
all patients who underwent CMR (22.8%), without significant differences between patients
with and without LVSD as per our classification (Table 3). Presence of LGE was observed in
22 patients of those with prior coronary artery disease (22%) and in 40 patients with pre-
existing hypertension (40%). Quantification of LVEF by CMR and TTE was strongly
positively correlated (r = 0.796, p < 0.01).
Patient flow through the clinic and rates of myocardial toxicity by referral stream are
summarized in Figure 1. Patients were mostly referred for optimisation before cancer
treatment (n=238, 44.4%, of whom 66.3% [156/238] were treatment naïve), and for
evaluation of LVSD or congestive heart failure (CHF) during or after cancer treatment
(n=134, 25.0%). Of the 535 patients, 347 (64.5%) had received any prior medical cancer
therapy and 136 patients (25.4%) had prior radiotherapy, of which only 37 (6.9%) received
potentially cardiotoxic radiotherapy to the left breast, left chest or mediastinum.
Patients at High Baseline Risk before Cancer Treatment
Two hundred and thirty-eight patients (44.4%) were deemed at high baseline risk by their
Oncology team and were referred for assessment prior to receiving further cancer surgery
(131/238, 55%) or potentially cardiotoxic first line cancer therapy (107/238, 45%). Median
LVEF was 61% (IQR 55 – 66%) but there was a high prevalence of CV risk factors in this
group including hypertension (42%) pre-existing CHF (10%) and established coronary artery
disease (CAD) (18%) underlying the clinical rationale for their referral. In these patients, we
detected an abnormal LVEF (less than 55%) in 18% and a significant structural abnormality
in a further 18% (including valvular heart disease, previous myocardial infarction,
hypertrophic cardiomyopathy or pericardial effusion). We performed stress
echocardiography in 157 of these 238 referred patients (66%) and inducible ischaemia was
8
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
identified in only 2%, who were subsequently referred for coronary angiography/ optimisation
of medical therapy. After our initial assessment, 110 patients (45.1%) were treated with beta-
blockers, 77 patients (31.6%) with ACE-inhibitors and 34 patients (13.6%) with Angiotensin II
receptor blockers. At this point, 86% of patients were deemed fit for further cancer treatment,
only 4% of patients were deemed too high risk and 10% of patients required optimisation.
Finally, of the latter patients, two-thirds successfully received their indicated cancer
treatment after optimisation, resulting in a final 92.4% cancer treatment rate (Figure 1a).
Rates of Myocardial Toxicity by the Royal Brompton Hospital Definition
Of the 347 patients who had previously been or were currently treated with any cardiotoxic
cancer therapy, 64.0% (n = 222) had evidence of cardiotoxicity by our classification (Table
4). Risk factors were common with 38.3% of patients having at least one pre-existing CV risk
factor. Most events in our series were symptomatic or asymptomatic LVEF reduction, which
were seen in 36.3% of patients (n = 126). Asymptomatic rises in BNP were also common
and seen in 14.4% of patients (n = 50) while isolated early functional changes with normal
LVEF were less common, occurring in 1.6% of patients (n = 6). Across all treatments, mixed
functional and biochemical toxicity was relatively common (9.7%, n = 34), whereas
symptomatic HFpEF was rare (1.6%, n = 6).
We concentrated our analysis to the most recognized cardio-toxic agents - anthracyclines,
anti-HER2 agents, VEGF TKIs and immune therapies which are detailed in Table 4.
Anthracyclines were the most commonly used cardio-toxic agents with the highest rate of
myocardial cardiotoxicity (78.0%) (n = 103). Total rates of cardiotoxicity were similar
between anthracyclines and anti-HER2 agents but symptomatic left ventricular systolic
dysfunction was less common in the anti-HER2 treated patients (28.8% vs 23.8%
respectively). Rates of cardiovascular toxicity in TKIs were significant, occurring in 65.2% of
patients (n = 43), driven primarily by hypertension, with direct myocardial toxicity reflected
mainly by a combination of symptomatic (22.7%, n=15) and asymptomatic (19.7%, n=13)
reduction in LVEF. ECGs from 30 patients treated with TKIs were available for analysis. The
9
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
median QTc was 439ms (Range 383 – 520ms), of whom 12 patients had a QTc greater than
450ms. Immunotherapies also had a high rate of cardiac toxicity (77.2%, n = 17), but with
relatively lower rates of reduction of LVEF (36.4%, n = 8), compared to other agents. These
proportions reflect the referred population which is influenced by development of new
cardiac symptoms on treatment and cardiac surveillance (anti-HER2 therapies).
Management and Outcomes of Patients with Myocardial Toxicity Requiring Ongoing
Cancer Treatment
Table 5 focuses on the patients with myocardial toxicity as per our criteria who required
ongoing medical cancer treatment irrespective of referral source. Across all the groups, rates
of optimisation of CV agents after our first assessment was high (125 patients, 73.1%), with
a progressive stepwise increase from Groups 1 – 6. The majority of patients were deemed
CV fit to continue with their cancer therapies (160 patients, 93.6%) including all patients in
Groups 1 - 4. Despite this, final cancer treatment completion rate was far lower due to
oncological reasons (112 patients, 65.5%) and survival at follow-up was poor in general (103
patients, 60.2%), and worst in the Group 1 of myocardial toxicity. CV and unexpected
mortality rates were generally low across all groups (14 patients, 20.6%).
Diagnosis, Management and Outcomes in the subgroup with LV dysfunction
One-hundred twenty-eight patients had evidence of LVSD from any cause (Table 3). The
primary cause of LVSD was chemotherapy in 104 patients (81%), with the remainder
including secondary to CAD, dilated cardiomyopathy and Takotsubo Syndrome. After the
initial assessment, 72% of patients were deemed suitable for initiation or continuation of
cancer therapy; after cardiac optimisation, this figure rose to 89% of patients (Figure 1b). As
a result, 67% of the patients with LVSD were able to complete their cancer drug regimens;
22% were deemed fit for drug therapy but did not ultimately receive further treatment after
discussion with the Oncology team; only 11% were considered too high risk from a
10
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
cardiovascular perspective. After the first appointment, 113 patients (88%) had their CHF
medications optimised as per current guidelines. Including follow-up, 120 patients (94%) had
an uptitration of their cardiac medications by our service. Table 6 presents the rates of most
frequently used CV agents after the first appointment and at follow-up.
Our follow-up rate was 88%, with a median follow-up of 360 days (IQR 84-735 days).
Median LV ejection fraction on assessment was 45% (IQR 0.39-0.50), which rose to 53%
(IQR 0.48-0.60) at follow-up (p <0.001). Overall, patients were less symptomatic at follow-
up, with NYHA class 3-4 in 27 patients (22%) on first assessment and 11 patients (10%) at
follow-up (p = 0.01). Mortality at follow-up was lowest in those who were deemed fit for
continuation of cancer treatment (16/66; 24.2%), higher in those CV suitable but ultimately
did not receive cancer treatment (11/22; 50.0%) and highest in those who were deemed too
high risk (8/10; 80%) (p < 0.01). In total, forty-four patients with LVSD (34.4%) died at follow-
up, with the cause of death being CV or unexpected in 10 patients, 22.7%. This death rate
was comparable to that in patients without LVSD (125 patients, 30.7%, of which the cause of
death was CV or unexpected in 26, 20.8%) (p=0.47). BNP levels at follow-up were available
in 79 patients (61.7%), with no significant difference with baseline (median 47, IQR 24-
117.75, p = 0.425).
Service Activity and Patient Feedback
For the period 2012 - 2016, clinic activity increased from 105 (in 2012) to 179 patients/ year
(in 2016) with a significant increase in the average number of new daycase assessment
patients seen in our weekly clinic (2.3+/-1.08 vs 3.64+/-1.12 patients/day, p < 0.001). For
the same period, median waiting time was 12 days [IQR 7-20]. The degree of satisfaction for
the overall daycase experience (1=extremely dissatisfied; 10=extremely satisfied) was high
with an average rating of 9.42.
DISCUSSION
11
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
The main finding of this 5 year real world experience of the first purpose-designed Cardio-
Oncology Service in the United Kingdom is that implementation of standardised protocols of
care can achieve optimisation of the cardiovascular status of patients at high baseline risk or
with established myocardial toxicity, leading to high levels of cancer treatment. There is a
signal that, in the highest risk patients with LVSD, cardiovascular optimisation and
continuation of cancer treatment may be associated with improved survival.
The rising prevalence of cardiovascular risk factors and cardiotoxicity in cancer has led to an
increasing demand for cardio-oncology services. Small scale reports (8), society guidelines
(9, 10) and more recently a position report from the European Society of Cardiology (ESC)
(3) have proposed surveillance and treatment strategies for patients at high baseline risk and
with established myocardial toxicity. One of the principal aims of a cardio-oncology service is
to improve LVEF and functional class which may then render patients more suitable for
ongoing cancer therapies, which is particularly important since there is emerging evidence
that interruptions in cancer treatment are associated with a higher incidence of cancer
recurrence (11). On the other hand, it is also conceivable, since mortality for many cancers
remains high, that optimisation of LVEF and functional class may not confer significant
benefit on prognosis. Certainly, in our cohort, CV and unexpected mortality rates were low,
reflecting the poor prognosis of cancer in these highly selected group of patients. There
remains limited data pertaining to the delivery and outcomes of cardio-oncology services,
particularly with respect to how they may impact cancer treatment and survival.
Previous randomised controlled trial data have shown that institution of CHF treatments at
baseline in high risk patients exposed to cardio-toxic treatments can reduce symptomatic
CHF and LVEF reduction but with an unclear effect on cancer treatment (12-15). Our study
shows that with cardiovascular optimisation of this group of patients, we were able to
optimise and deem fit from a CV point of view the majority of patients referred before
continuing cancer treatment, who may not have experienced delays or had significantly
lower treatment rates. We developed an assessment tool for referring hospitals to identify
patients at high baseline risk and we follow the current ESC guidelines for non-cardiac
12
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
surgery as a foundation, but with an individualised approach to the patient (e.g. for neo-
adjuvant chemotherapy with cardio-toxic drugs) (16). The patients in our series had a
significant burden of cardiovascular risk factors but the low rates of ischaemia observed are
likely to reflect possible separate referral pathways for patients with ischaemic symptoms,
e.g. direct to interventional cardiologists.
The most commonly used definition for myocardial toxicity is the CREC [low LVEF (≤50%) or
significant drop of LVEF (>10%) below 55% during chemotherapy with or without
symptoms], and rates of myocardial toxicity in clinical trials have been less than 5% (17).
Rates of myocardial toxicity in clinical trials have varied; this reflects the baseline risk of
recruited patients, short durations of follow-up and variability of definitions of myocardial
toxicity for which there is no clear consensus (18, 19). We designed the Royal Brompton
Hospital Classification of Cancer Therapy-induced Myocardial Toxicity to incorporate early
abnormalities since they may predispose to future susceptibility, identify higher-risk patients
that could benefit from pre-emptive therapy and since it is more applicable to current clinical
practice. Using this classification, the rate of established myocardial toxicity in our series was
significantly higher than in clinical trials and previously reported literature (17). The increased
rates of myocardial toxicity observed in our series reflects a selected population of higher
risk and symptomatic patients, who may have been excluded from clinical trials owing to
high rates of prevalent CV disease, as well as the inclusion of biochemical and sub-clinical
markers of dysfunction. Consistent with other registries (20), rates of asymptomatic
reductions in LV ejection fraction were high reflecting appropriate referral of those with high
baseline risk driven by pre-existing cardiovascular disease and those who had cardiotoxicity
detected during standard surveillance protocols in the oncology services e.g. LVSD in
patients receiving trastuzumab for HER2+ breast cancer (18, 21). The long-term clinical
significance of de novo asymptomatic LV diastolic dysfunction or biochemical or functional
cardiotoxicity with a preserved LVEF (>55%) is currently unclear, but as these findings are
an objective marker of myocardial toxicity, they have been included in our analysis. For
example, BNP elevation in the setting of normal LVEF may reflect increased myocardial wall
13
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
strain directly related to the cancer or treatment (22, 23). Indeed, in our study, despite high
levels of CV fitness for therapy and similar rates of cancer treatment, the cohort of patients
with isolated biochemical myocardial toxicity (Group 1) experienced the highest mortality
amongst all. This may indicate that subtle increases in BNP are a signal for an early but
clinically significant process of myocardial toxicity or perhaps may reflect more advanced
systemic disease. This could also explain why BNP values did not differ between different
the subgroups of patients and did not change over time. Future studies evaluating the
significance of these findings, effect on outcome and rationale for treatment are required.
We specifically focused on patients with reduced LVEF, as this has been an exclusion
criteria in most clinical trials involving potentially cardio-toxic drugs. A previous small
retrospective study suggested that for patients with established myocardial toxicity, a
cardiology consultation was associated with up-titration of heart failure medications and
improved survival (24). Our protocolised model of care led to improvements in LVEF and
NYHA class in accordance with previous studies (25, 26) but also, importantly, a high rate of
continuation of cancer therapies. This subset of patients was younger and with fewer
comorbidities than patients without LVSD, but with a higher rate of oncological treatments.
Although the numbers were small, we provide preliminary evidence that being deemed
cardiovascularly fit and continuing cancer therapy specifically in patients with LVSD may
reduce mortality.
After evaluation in our service, all patients without LVSD were deemed CV fit but cancer
treatment completion rates were significantly lower due to oncological reasons. Mortality
rates were generally comparable across all groups of myocardial toxicity and worst in Group
1 and with similar rates of CV death. This highlights that the decision to continue cancer
treatment is complex and emphasises the importance of a multi-disciplinary process. We
could, however, hypothesise that with appropriate CHF treatment and improved LVEF in
patients with LVSD particularly, we rendered patients more able to complete their
14
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
oncological treatment – hence improving outcome. A prospective, larger scale comparison of
Cardio-Oncology services with current standard of care is required to clarify this issue.
We operate a one-stop day case model in order to provide a rapid opinion to the referring
unit thereby minimising interruptions of care and to improve the patient experience which
was confirmed by the excellent feedback we received. This patient group has, by definition,
received a double blow, firstly that of a cancer diagnosis, which itself is challenging, but then
compounded by actual or threatened heart disease. The latter discovery can be
psychologically challenging for patients, but the support gained from a team with experience
in this specific area and with treatments to offer had major positive impact.
Study Limitations
This descriptive analysis represents a clinical experience from a single centre. Although this
allows a uniform protocol for assessment and treatment, it is therefore susceptible to a
referral bias based on the expertise of our main referral centre. We recognise that patients
with LVSD may have been younger due to a referral bias from breast cancer and sarcoma
services and that the higher rates of cardiotoxicity reported are likely to represent a higher
risk case mix referred to our service. Certain malignancies were not commonly seen in this
service, for example acute leukaemias, but patients with 33 types of cancer were
represented, it is likely that the results are generalisable. We also acknowledge that there
was no control group of patients at high baseline risk who did not receive optimisation, but
since without optimisation their treatment rates would have certainly been much lower, we
feel the high rates of cancer treatment continuation remain relevant. Finally, similarly, a
comparator group of patients with LVSD not receiving cardiac treatment was not included
due to the specific nature of our service but future studies should consider comparing
outcomes of cardio-oncology services with standard care.
Conclusions
15
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
In our descriptive analysis reporting the five year activity of a UK based Cardio-Oncology
service, we observed a higher rate of prior CV disease and myocardial toxicity than that
previously reported in the literature. Prompt evaluation of CV risk with optimisation led to
improvements in LVEF and NYHA class and high rates of cancer treatment continuation.
References
1. Ameri P, Canepa M, Anker MS, Belenkov Y, Bergler-Klein J, Cohen-Solal A, et al. Cancer diagnosis in patients with heart failure: epidemiology, clinical implications and gaps in knowledge. Eur J Heart Fail. 2018.2. Maddams J, Utley M, Moller H. Projections of cancer prevalence in the United Kingdom, 2010-2040. Br J Cancer. 2012;107(7):1195-202.3. Zamorano JL, Lancellotti P, Rodriguez Munoz D, Aboyans V, Asteggiano R, Galderisi M, et al. 2016 ESC Position Paper on cancer treatments and cardiovascular toxicity developed under the auspices of the ESC Committee for Practice Guidelines: The Task Force for cancer treatments and cardiovascular toxicity of the European Society of Cardiology (ESC). Eur J Heart Fail. 2017;19(1):9-42.4. Suter TM, Ewer MS. Cancer drugs and the heart: importance and management. Eur Heart J. 2013;34(15):1102-11.5. Mulrooney DA, Yeazel MW, Kawashima T, Mertens AC, Mitby P, Stovall M, et al. Cardiac outcomes in a cohort of adult survivors of childhood and adolescent cancer: retrospective analysis of the Childhood Cancer Survivor Study cohort. BMJ. 2009;339:b4606.6. Patnaik JL, Byers T, DiGuiseppi C, Dabelea D, Denberg TD. Cardiovascular disease competes with breast cancer as the leading cause of death for older females diagnosed with breast cancer: a retrospective cohort study. Breast Cancer Res. 2011;13(3):R64.7. Cardinale D. [A new frontier: cardio-oncology]. Cardiologia. 1996;41(9):887-91.8. Okwuosa TM, Barac A. Burgeoning Cardio-Oncology Programs: Challenges and Opportunities for Early Career Cardiologists/Faculty Directors. J Am Coll Cardiol. 2015;66(10):1193-7.9. Curigliano G, Cardinale D, Suter T, Plataniotis G, de Azambuja E, Sandri MT, et al. Cardiovascular toxicity induced by chemotherapy, targeted agents and radiotherapy: ESMO Clinical Practice Guidelines. Ann Oncol. 2012;23 Suppl 7:vii155-66.10. Plana JC, Galderisi M, Barac A, Ewer MS, Ky B, Scherrer-Crosbie M, et al. Expert consensus for multimodality imaging evaluation of adult patients during and after cancer therapy: a report from the American Society of Echocardiography and the European Association of Cardiovascular Imaging. Eur Heart J Cardiovasc Imaging. 2014;15(10):1063-93.11. Ye et al. Impact of Trastuzumab-Induced Cardiotoxicity and Subsequent Trastuzumab Interruption on Breast Cancer Outcome. American College of Cardiology Annual Congress2013.12. Kalay N, Basar E, Ozdogru I, Er O, Cetinkaya Y, Dogan A, et al. Protective effects of carvedilol against anthracycline-induced cardiomyopathy. J Am Coll Cardiol. 2006;48(11):2258-62.13. Georgakopoulos P, Roussou P, Matsakas E, Karavidas A, Anagnostopoulos N, Marinakis T, et al. Cardioprotective effect of metoprolol and enalapril in doxorubicin-treated lymphoma patients: a prospective, parallel-group, randomized, controlled study with 36-month follow-up. Am J Hematol. 2010;85(11):894-6.14. Bosch X, Rovira M, Sitges M, Domenech A, Ortiz-Perez JT, de Caralt TM, et al. Enalapril and carvedilol for preventing chemotherapy-induced left ventricular systolic dysfunction in patients with malignant hemopathies: the OVERCOME trial (preventiOn of left Ventricular dysfunction with Enalapril and caRvedilol in patients submitted to intensive
16
1
2
3
4
5
6789
1011121314151617181920212223242526272829303132333435363738394041424344454647484950
ChemOtherapy for the treatment of Malignant hEmopathies). J Am Coll Cardiol. 2013;61(23):2355-62.15. Gulati G, Heck SL, Ree AH, Hoffmann P, Schulz-Menger J, Fagerland MW, et al. Prevention of cardiac dysfunction during adjuvant breast cancer therapy (PRADA): a 2 x 2 factorial, randomized, placebo-controlled, double-blind clinical trial of candesartan and metoprolol. Eur Heart J. 2016;37(21):1671-80.16. Kristensen SD, Knuuti J, Saraste A, Anker S, Botker HE, Hert SD, et al. 2014 ESC/ESA Guidelines on non-cardiac surgery: cardiovascular assessment and management: The Joint Task Force on non-cardiac surgery: cardiovascular assessment and management of the European Society of Cardiology (ESC) and the European Society of Anaesthesiology (ESA). Eur Heart J. 2014;35(35):2383-431.17. Ewer MS, Ewer SM. Cardiotoxicity of anticancer treatments. Nat Rev Cardiol. 2015;12(11):620.18. Slamon D, Eiermann W, Robert N, Pienkowski T, Martin M, Press M, et al. Adjuvant trastuzumab in HER2-positive breast cancer. N Engl J Med. 2011;365(14):1273-83.19. Eschenhagen T, Force T, Ewer MS, de Keulenaer GW, Suter TM, Anker SD, et al. Cardiovascular side effects of cancer therapies: a position statement from the Heart Failure Association of the European Society of Cardiology. Eur J Heart Fail. 2011;13(1):1-10.20. Farolfi A, Melegari E, Aquilina M, Scarpi E, Ibrahim T, Maltoni R, et al. Trastuzumab-induced cardiotoxicity in early breast cancer patients: a retrospective study of possible risk and protective factors. Heart. 2013;99(9):634-9.21. Tocchetti CG, Ragone G, Coppola C, Rea D, Piscopo G, Scala S, et al. Detection, monitoring, and management of trastuzumab-induced left ventricular dysfunction: an actual challenge. Eur J Heart Fail. 2012;14(2):130-7.22. Pavo N, Raderer M, Hulsmann M, Neuhold S, Adlbrecht C, Strunk G, et al. Cardiovascular biomarkers in patients with cancer and their association with all-cause mortality. Heart. 2015;101(23):1874-80.23. Daugaard G, Lassen U, Bie P, Pedersen EB, Jensen KT, Abildgaard U, et al. Natriuretic peptides in the monitoring of anthracycline induced reduction in left ventricular ejection fraction. Eur J Heart Fail. 2005;7(1):87-93.24. Ammon M, Arenja N, Leibundgut G, Buechel RR, Kuster GM, Kaufmann BA, et al. Cardiovascular management of cancer patients with chemotherapy-associated left ventricular systolic dysfunction in real-world clinical practice. J Card Fail. 2013;19(9):629-34.25. Cardinale D, Colombo A, Lamantia G, Colombo N, Civelli M, De Giacomi G, et al. Anthracycline-induced cardiomyopathy: clinical relevance and response to pharmacologic therapy. J Am Coll Cardiol. 2010;55(3):213-20.26. Cardinale D, Colombo A, Sandri MT, Lamantia G, Colombo N, Civelli M, et al. Prevention of high-dose chemotherapy-induced cardiotoxicity in high-risk patients by angiotensin-converting enzyme inhibition. Circulation. 2006;114(23):2474-81.
17
123456789
101112131415161718192021222324252627282930313233343536373839
40
41
42
Figure Legends
Figure 1. Diagrams of Patient Flows after Assessment and Optimisation By Our
Service
Panel A. Diagram of decision making for patients referred for cardiovascular optimisation at
baseline before current cancer treatment Panel B. Diagram of decision making and outcome
in patients with LVSD (groups 5 and 6 of our cardiac dysfunction classification) awaiting
potential cardiotoxic therapy.
Table 1. Reasons for Referral to Cardio-Oncology Service, Baseline Assessment and Follow-upCondition Baseline Assessment Follow-up
High baseline CV risk pre-operation or pre-cancer
therapy
Resting 2d TTE. CMR and/ or stress TTE if clinically appropriate.
ESC Guidelines for non-cardiac surgery as foundation. If LVSD, as approach for
cancer therapy as in Table xx.
Clinically indicated.If low-risk – discharge
Consider referral to sub-specialities – Interventional Cardiology or
Electrophysiology
Post-cancer asymptomatic or symptomatic therapy
LVSD (AC, HER-2, TKI and immunotherapy)
Resting 2d TTE and biomarker assessment. CMR and/ or stress TTE if
appropriate.Guideline based HF therapy or
personalised treatment in selected cases.
TTE & biomarkers 4 weekly to 3 monthly. Intervals depend on
severity of LVSD, requirement of therapy, cancer stage and
response
Vasospasm secondary to chemotherapy
Resting 2d TTE, stress echocardiography or coronary
angiography with ergotamine infusion. Treat with vasodilators and consider
intra-venous GTN infusion during chemotherapy administration
Routine follow-up
VEGF-inhibitor therapy – Prolonged QTc or
Hypertension
Ambulatory BP, resting 2d TTE and CMR if appropriate. 12 lead ECG and EP
consult if indicated. Commence guideline-based anti-hypertensive
treatment.
TTE 3 monthly with home BP diary. Supported 12 ECG monitoring in
local centre.
Evaluation of cardiac tumours
Resting 2d TTE, Biomarkers, CMR with EGE and LGE. Liaison with
Interventional Cardiology/ CTS for Biopsy or resection if indicated
Interval TTE or CMR to evaluate response to treatment
Direct cardiac complications of cancer – e.g. carcinoid
valvular heart disease, cardiac amyloidosis,
pericardial effusion, direct invasion
Resting 2d TTE, Biomarkers, CMR with specialised protocols. Consider referral
to other specialties e.g. amyloidosis centres, interventional cardiology or CTS
for pericardial or valvular surgery
Personalised follow-up based on treatment, cancer therapy, risk of
embolism or stage of disease (e.g. moderate valvular disease)
18
1
2
3
4
5
6
7
Table 2. Management strategies and percentage of patients completing cardiotoxic cancer therapy according to myocardial toxicity class
Group
Classification
Definition
Management
strategiesOncology
tx
Management strategiesCardiology tx
1Biochemical dysfunction
New BNP or Troponin I rise but with normal cardiac imaging
Continue Cardio-Oncology reviewConsider closer monitoring or start lower dose ACE I or
BB cardioprotection
2Early functional dysfunction
New reduction in GLS or grade III-IV diastolic dysfunction] and normal biomarkers
Continue Cardio-Oncology reviewConsider closer monitoring or start lower dose ACE I or
BB cardioprotection
3Mixed dysfunction
Normal LVEF with abnormal biomarkers and GLS/diastolic dysfunction
Continue Cardio-Oncology reviewStart lower dose ACE I or
BB cardioprotection
4 HFpEF
Symptomatic heart failure with preserved ejection fraction
Interrupt and review risk:benefit*
Cardio-Oncology reviewDiuretic for fluid congestion
ACE I or BB cardioprotection if
continuing cancer tx
5Asymptomatic LVSD
New LVEF reduction to <50% or a reduction in LVEF >10% to a LVEF <55%)
Review and balance
risk:benefit*
Cardio-Oncology reviewStart ACE I and/or BB and uptitrate to 50-100% target dose for HF as tolerated#
6 HFrEF
Symptomatic reduction in LVEF <50% or a reduction in LVEF >10% to a LVEF <55%).
Interrupt and review risk:benefit*
Cardio-Oncology reviewStart ACE I and/or BB and
uptitrate to 100% target dose for HF as tolerated#**
*Continuing cardiotoxic cancer therapy may be suitable in selected cases depending risk:benefit, severity of LV impairment, symptoms, cancer stage and response
# If ACE I or BB not tolerated, or patient already taking these agents when cardiotoxicity diagnosed consider adding aldosterone antagonist
19
1
2
3
45
67
** If LVEF<35% follow ESC HF guideline regarding eligibility for CRT, sacubitril/valsartan and ivabradine
Table 3. Baseline Patient CharacteristicsAll
(n=535)No LVSD (n=407)
LVSD (n=128)
P
Age 60.5 +/-
14.7
62.0 +/- 14.3 55.9 +/- 15.4 p < 0.001
Female n (%) 299 (55.8) 219 (53.5) 80 (62.5) p = 0.074
CV risk factors n (%)Hypertension
Previous CHF
Valvular Heart Disease
Prior CAD
181 (33.8)
33 (6.1)
27 (5.0)
54 (10.1)
155 (37.9)
25 (6.1)
25 (6.1)
47 (11.5)
26 (20.3)
8 (6.25)
2 (1.6)
7 (5.5)
p < 0.001p = 0.697
p = 0.043p = 0.070
Chemotherapy n (%)Anthracycline
Anti-HER2*
TKI†
Agents causing Vasospasm+
347 (64.7)
132 (24.6)
63 (11.8)
66 (12.3)
93 (17.3)
219 (53.5)
59 (14.4)
27 (6.6)
36 (8.8)
54 (13.2)
128 (100)
73 (57)
36 (28.1)
30 (23.4)
39 (30.5)
p < 0.001p < 0.001p < 0.001p < 0.001p < 0.001
Previous Radiotherapy n (%)
139 (25.9) 95 (23.2) 44 (34.4) p = 0.013
Previous Surgery n (%) 252 (47.0) 173 (42.3) 79 (61.7) p < 0.001Median LVEF (%) (IQR) 60 (53 – 65) 62 (59 – 67) 47 (40 – 53) p < 0.001BNP positive, n (%) 80 (16.5) 60 (16.2) 20 (17.5) p = 0.739
BNP (ng/L) [Median (IQR)] 63 (34 -
147)
57 (33;124) 96.5 (37;
225)
p = 0.564
Troponin positive, n (%) 27 (6) 15 (4.5) 12 (11) p = 0.013CMR LGE, n (%) 100 (23.4) 75 (23.2) 25 (23.8) p = 0.901
20
12
3
4
Percent and raw data in brackets. Cardiac dysfunction defined as per our proposed classification* Trastuzumab, Pertuzumab & Lapatinib† Imatinib, Sorafenib, Sunitinib, Pazoponib, Axitinib, Vandetanib, Vemurafenib, Gefitinib, Lenvatinib, Cobimetinib+ 5-FU and Capecitabine
21
1234567
8
22
Table 4. Rates of Cancer Therapy-induced Myocardial ToxicityAll Anthrac
yclineAnti-HER2
TKI Immunotherapy
Total - n 347 132 63 66 22
Age [Median (IQR)] 59 (21) 51 (20) 54 (17) 59 (18) 70 (18.17)
Risk factors - n (%) Hypertension Previous CHF Valvular disease Prior CAD
92 (26.5)15 (4.3)10 (2.9)16 (4.6)
21 (15.9)3 (2.3)3 (2.3)2 (1.5)
12 (19.0)1 (1.6)2 (3.2)
0
22 (33.3)1 (1.5)1 (1.5)4 (6.1)
7 (31.8)0 (0)1 (0)0 (0)
Cancer therapies - n (%) Anthracylines Anti-HER2 TKI
132 (38.0)
63 (18.2)66 (19.0)
N/A39 (29.5)13 (9.8)
39 (61.9)N/A
4 (6.3)
13 (19.7)4 (6.0)
N/A
5 (22.7)0 (0)
4 (18.2)
Immunothrapy 22 (6.3) 5 (3.8) 0 (0) 4 (6.0) N/A
LVEF (%) [Median (IQR)] 58.2 (13) 55 (15.0) 56
(13.9)
57 (15) 58.5 (12.5)
BNP positive - n (%)BNP (ng/L) [Median (IQR)]Troponin positive - n (%)LGE – n (%)
250
(72.0)
58
(117.3)
21 (70.9)
60 (17.3)
88 (66.7)
49 (133)
17 (12.9)
20 (15.2)
37
(55.6)
37 (88)
2 (3.2)
7 (11.1)
47 (71.2)
46 (86)
2 (3.0)
8 (14.8)
19 (86.4)
91 (84.5)
1 (4.5)
6 (27.3)
Rates of Cancer Therapy-induced Myocardial Toxicity
Any cardiotoxicity n (%)222
(64.0)103
(78.0)46
(73.0)43 (65.2) 17 (77.2)
Biochemical cardiotoxicity - n (%)
50 (14.4) 14 (10.6) 7 (11.1) 9 (13.6) 4 (18.1)
Early Functional cardiotoxicity - n (%)
6 (1.6) 3 (2.3) 1 (1.6) 0 (0) 1 (4.5)
Mixed Early cardiotoxicity - n (%)
34 (9.7) 9 (6.8) 5 (8.0) 5 (7.6) 4 (18.1)
Symptomatic HFpEF - n (%)
6 (1.6) 3 (2.3) 0 (0) 1 (1.5) 0 (0)
Asymptomatic LVSD - n (%)
53 (15.2) 36 (27.3)18
(28.6)13 (19.7) 3 (13.6)
Symptomatic LVSD - n (%)
73 (21.0) 38 (28.8)15
(23.9)15 (22.7) 5 (22.7)
1
Table 5. Management and survival of patients with myocardial toxicity Requiring Pharmacological Cancer Treatment.
GroupCV
treatment optimised –
n (%)
CV fit for cancer treatment – n
(%)
Completed cancer
treatment – n (%)
Survival – n (%)
CV or Unexpected mortality - n (%)
1n = 39 18 (46.2) 39 (100) 26 (66.7) 21 (53.8) 3 (16.7)
2n = 5
3 (60.0) 5 (100) 4 (80.0) 3 (60.0) 0 (0)
3n = 28
18 (64.3) 28 (100) 20 (71.1) 18 (64.3) 2 (20.0)
4n = 3
2 (66.7) 3 (100) 1 (33.3) 2 (66.7) 1 (100)
5n = 43
36 (83.7) 40 (93.0) 27 (62.8) 27 (62.8) 3 (18.8)
6n = 53
48 (90.6) 45 (84.9) 34 (64.2) 32 (60.4) 5 (23.8)
Total, n = 171 125 (73.1) 160 (93.6) 112 (65.5) 103 (60.2) 14 (20.6)
23
1
2
3
4
Table 6. Rates and dosage of heart failure agents in patients with LVSD
AgentAfter first
appointment - n(%)
Equivalent dose – mg, median
(IQR)At follow up – n
(%)
Follow-up equivalent dose
– mg, median (IQR)
Betablockers 91 (71.1) 2.5 (1.25-2.5) 99 (77.3) 2.5 (2.5-5)
ACEI/ARB 104 (81.3) 2.5 (1.25-2.5) 108 (84.4) 2.5 (2.34-5.63)
MRA 17 (13.3) 25 (25-25) 27 (21.1) 25 (25-50)
Diuretics 28 (21.7) N/A 31 (24.2) N/A
ACEI: ACE Inhibitors; ARB: Angiotensin II Receptor Blockers; MRA Mineralocorticoid receptor antagonist
Equivalent doses were calculated for 24h dosage of Bisoprolol, Ramipril and Spironolactone, respectively
N/A: not applicable
24
12
3
45
6
