British Journal of Cancer www.nature.

com/bjc

REVIEW ARTICLE OPEN

Clinical Studies

Cardiotoxicity following thoracic radiotherapy for lung cancer

✉
Gerard M. Walls 1,2 , Carmen Bergom1,3, Joshua D. Mitchell 4
, Stacey L. Rentschler5,6, Geoffrey D. Hugo1,3, Pamela P. Samson1,3 and
1,3
Clifford G. Robinson

© The Author(s) 2024

Radiotherapy is the standard of care treatment for unresectable NSCLC, combined with concurrent chemotherapy and adjuvant
immunotherapy. Despite technological advances in radiotherapy planning and delivery, the risk of damage to surrounding thoracic
tissues remains high. Cardiac problems, including arrhythmia, heart failure and ischaemic events, occur in 20% of patients with lung
cancer who undergo radiotherapy. As survival rates improve incrementally for this cohort, minimising the cardiovascular morbidity
of RT is increasingly important. Problematically, the reporting of cardiac endpoints has been poor in thoracic radiotherapy clinical
trials, and retrospective studies have been limited by the lack of standardisation of nomenclature and endpoints. How baseline
cardiovascular profile and cardiac substructure radiation dose distribution impact the risk of cardiotoxicity is incompletely
understood. As Thoracic Oncology departments seek to expand the indications for radiotherapy, and as the patient cohort becomes
1234567890();,:

older and more comorbid, there is a pressing need for cardiotoxicity to be comprehensively characterised with sophisticated
oncology, physics and cardio-oncology evaluations. This review synthesises the evidence base for cardiotoxicity in conventional
radiotherapy, focusing on lung cancer, including current data, unmet clinical needs, and future scientific directions.

British Journal of Cancer; https://doi.org/10.1038/s41416-024-02888-0

BACKGROUND use, lifestyle factors and social deprivation [10]. This review
Lung cancer is consistently ranked within the highest causes of synthesises the evidence base for cardiotoxicity in conventionally
death in Western populations [1]. With its high incidence and fractionated RT specifically and focuses on lung cancer via
mortality [2], it is recognised as a ‘cancer of unmet need’ [3]. established research findings, unmet clinical needs, and future
Definitive radiotherapy (RT) combined with systemic therapy offers directions.
the greatest probability of long-term disease control for the high
proportion of patients who have potentially curable disease, where
surgery is not possible. However, definitive RT carries cardiovascular CARDIOTOXICITY OF RADIOTHERAPY: HISTORICAL CONTEXT
risk and minimising the morbidity of this treatment is paramount. The first scientific descriptions of radiation-induced heart disease
Cancer control rates from RT are also suboptimal given that 25–40% (RIHD) were made in 1924. A German pathologist found extensive
of patients will experience local relapse [4], yet an increased RT dose cardiac scar formation during the autopsy of a patient previously
led to worse overall survival (OS) in the landmark dose-escalation treated with RT for mediastinal lymphoma who subsequently died
trial, RTOG-0617 [5]. The elevated mortality rate was associated with from heart failure [11]. A canine experiment investigating
higher cardiac radiation doses in this trial. This finding was radiation-related pulmonary effects in the same year reported
corroborated by the recent phase 3 LungART trial of postoperative fibrotic changes in the atria [12]. However, RIHD was regarded as a
lung RT, where improved disease control with RT likely did not rare sequela for a further 4 decades until a case series of breast
translate into improved OS due to deleterious cardiopulmonary and lymphoma RT cases implored that the heart was not as
effects [6]. Pooled analyses of historical NSCLC radiation trials radioresistant as previously thought in 1967 [13]. Interest in RIHD
examining symptomatic post-RT cardiac events (PRCEs) identified a then emerged among lung radiation oncologists during 9
2-year cumulative incidence rate of 11–32% [7, 8], consisting largely adjuvant trials over the following 3 decades [14], although careful
of arrhythmia, heart failure and ischaemic events. Also relevant is attention to minimising heart doses was not embedded in routine
the high prevalence of cardiovascular risk factors (CVRFs) and practice given the prioritisation of improving dismal cancer
established cardiac diseases (ECDs) (Fig. 1) before treatment that is control rates in lung cancer, and the lack of technology adequate
typical of lung cancer cohorts [9], owing to considerable tobacco to do so in most of the older trials.

1
Department of Radiation Oncology, Washington University in St Louis, Saint Louis, MO, USA. 2Patrick Johnston Centre for Cancer Research, Queen’s University Belfast, Belfast,
Northern Ireland, USA. 3Siteman Cancer Center, Washington University Medical Campus, Saint Louis, MO, USA. 4Cardio-Oncology Center of Excellence, Washington University in
St Louis, St Louis, MO, USA. 5Department of Developmental Biology, Washington University in St Louis, St. Louis, MO, USA. 6Center for Cardiovascular Research, Department of
Medicine, Cardiovascular Division, Washington University in St Louis, St. Louis, MO, USA. ✉email: [email protected]

Received: 26 May 2024 Revised: 16 October 2024 Accepted: 21 October 2024

 G.M. Walls et al.
2
a
Pre-radiotherapy Post-radiotherapy

Cardiovascular Established Post-radiotherapy

risk factors cardiac diseases cardiac events
CVRFs ECDs PRCEs

Hypertension
Dyslipidaemia Heart failure Heart failure
Diabetes mellitus Arrhythmia Arrhythmia
Smoking Angina/MI Angina/MI
Family history

Other potentially relevant conditions

Cardiac – valvular disease, pericardial disease, cardiomyopathy
Vascular – cerebrovascular disease, peripheral vascular disease, aortic aneurysm
Genetic – familial hypercholesterolaemia, family history of IHD <65 years, connective tissue disorders

Endpoint Strengths Limitations

x Not specific to cardiotoxicity

x Reliable endpoint
x Difficult to determine cause of death
Overall x Straightforward to collect
x Dependent on patient factors
survival x Recognised by Oncologists
x Dependent on tumour factors
x Recognised by Cardiologists
x Dependent on treatment factors

x Cardiac-specific x Encompasses several substructures

Major adverse x Clinically meaningful
x Omits clinically significant arrhythmias
cardiac events x Higher event rate
x Multiple different definitions
x Recognised by Cardiologists

x Two closely linked pathologies

Ischaemia ± x Encompasses several substructures
x Clinically meaningful
heart failure x Increases the event rate

x Physiologically relevant
Single toxicity x Decreased event rate
x Promotes in-depth analysis
endpoints x Clinically meaningful

Fig. 1 Categorisation of cardiovascular factors and study endpoints in radiation heart disease particularly in the setting of lung cancer. a
is a summary schema for how critical cardiovascular factors can be considered temporally. b displays the positive and negative characteristics
of commonly employed study designs. (MI = myocardial infarction; IHD = ischaemic heart disease).

In 1991 a dedicated interdisciplinary working group in North radiosensitivity at the time, as excluding the blood pool from
America made recommendations on the tolerance of normal organs calculations did not alter results [21].
in partial and complete irradiation scenarios [15]. Emami et al. An initiative to summarise the normal tissue toxicity literature
focused on what was considered to be the most severe clinical two decades after the adoption of three-dimensional RT planning
endpoint for an organ at the time, which was pericarditis for the (3DCRT) was unable to make significant advancements for heart
heart. The authors highlighted that the lymphoma and breast constraints, simply concluding that the incidental cardiac dose
cancer-based recommendations for heart dose constraints, is “… should be as low as reasonably achievable (ALARA) [22]. The
mostly speculative…”, “…from sporadic information in the litera- authors noted that lung cancer data requires special considera-
ture…”, and involved the “…clinical impressions of the clinicians tion, since this patient cohort was not included in prior RIHD
involved…”. These suggested safe dose constraints remained the analyses and yet had a high rate of PRCEs. The duality of the
best available 16 years later, during the design of the RTOG-0617 heart’s radiobiological tissue organisation was also acknowledged,
study. To test the principles of ‘the Emami paper’, the diverse range with the left ventricular myocardium possibly being regarded as a
of dose-fractionations were converted into a common format, based ‘parallel’ tissue with respect to muscle contraction, but a ‘serial’
on a radiobiological characteristic, the α/β ratio [16], which animal tissue with respect to conduction. Exemplifying the important role
data suggested was 2.5–3.0 [17–19] for pericardium-related end- of RT technology in RIHD, intensity modulated RT (IMRT), a
points. The α/β ratio is a summary value for the linear-quadratic planning/delivery approach with greater conformality than 3DCRT
behaviour of an organ’s likelihood of developing a specific toxicity introduced in the 2000s, has routinely demonstrated lower
[20]. The heart was considered as a single structure with uniform maximum cardiac doses than 3DCRT [23].

British Journal of Cancer

 G.M. Walls et al.
3
CARDIOTOXICITY IN CLINICAL TRIALS calculating the equivalent dose in 2 Gy fractions, and image-
RIHD was not truly considered in lung cancer clinics until the guidance details were rarely provided. The mean number of
RTOG-0617 results were published in 2015. This phase 3, patients and follow-up duration across all studies were 334 (range
randomised trial in stage III NSCLC found OS to be worse in the 43–1190) and 32 months (range 13–74), and patients were treated
high-dose arm (74 Gy) at 20 months versus 29 months in the exclusively within trials in 10 studies.
standard-dose arm (60 Gy) [24]. In a multivariate analysis (n = 407) The primary endpoint was cardiac events for 17/42 studies and
the volume of heart receiving at least 5 Gy (V5) and 30 Gy (V30) OS in 20/42 studies, with other endpoints including electrocardio-
were significantly associated with reduced OS, and a secondary gram (ECG) changes (Table 1). Only a minority of studies reported
analysis indicated V40 was the best heart dose predictor [23]. A baseline cardiac status, i.e. CVRFs (e.g. hypertension, dyslipidae-
retrospective study (n = 416) by the RTOG-0617 chief investigators mia, diabetes mellitus, smoking, family history) and ECDs (e.g.
in 2016 corroborated that several heart dose metrics, including arrhythmia, heart failure (HF) and acute coronary syndrome (ACS)).
V50, are associated with reduced OS, and increased PRCEs, Cardiovascular risk scores were provided in only 10/42 studies.
establishing this clinical problem as a new priority [25]. A diverse range of cardiac dose constraints were used during
Cardiac outcomes were not specifically embedded as a study treatment planning, and where these were reported, they were for
endpoint in RTOG-0617, but acute events were captured via the whole heart (WH). Hearts were generally contoured based on
standard toxicity assessment. PRCEs occurred in 3% and 9% the Feng atlas [31] (n = 15/42) or Radiation Therapy Oncology
patients in the standard- and high-dose arms respectively, a rate Group guidance [32] (n = 6/42) (Table 1). Sixteen studies
comparatively lower than published retrospective studies. Ubiqui- delineated at least 1 cardiac chamber, and at least 1 coronary
tous conditions such as hypertension and sinus tachycardia were artery was contoured in 7 studies. Chamber walls, without the
excluded in these calculations, and it is noteworthy that the intraluminal blood pool, were segmented in 1 study [33]. The most
patients were fitter than the general population as eligibility common dose volume histogram (DVH) summary metrics
required Eastern Co-operative Oncology Group (ECOG) status of employed were the mean dose (Dmean) and maximum dose
0–1. Fatal adverse events were due to cardiac causes in 6 cases, (Dmax).
and possible cardiac causes (‘sudden death’, ‘death not otherwise The large range of average mean heart doses (MHD) among
specified’) in 6 cases, although cause of death data was not patients in the cohorts, 2.2–16.6 Gy, epitomises two decades of RT
reviewed by the central trial team. These values may also be technology evolution, the large variability in the anatomic
underestimates, given the difficulty of determining cause of death relationship of the lung and involved nodes relative to the heart,
in lung cancer [26]. There were considerably more non-protocol and global dose-fractionation prescription diversity (Table 1). The
contouring deviations for the heart ( ~ 20%) than any other organ- mean incidence of pooled PRCEs was 19.7% (range 7–32%) with
at-risk (OAR) (0–4%) reflecting how cardiac effects were of low an average follow-up of 26.0 months (range 13.2–45.0). Of the
concern when the trial activated in 2007 [24]. Furthermore, a 14 studies where a multivariate analysis was performed for a
breach of the recommended dose limits (V40 < 100%, V45 < 66%, dosimetric parameter as a predictor of PRCEs, 10 found a
V60 < 33%) was permitted to facilitate any other OAR meeting its statistically significant association. Using OS as a surrogate
threshold. endpoint for RIHD, 30 of 38 studies found statistically significant
Another landmark lung RT trial in recent years also exhibited an associations with at least one DVH metric on multivariate analysis.
OS detriment associated with cardiac irradiation. The LungART Only one study included patients that underwent curative-intent
trial randomised patients with fully resected stage N2 NSCLC to re-irradiation [34].
adjuvant RT (heart V30 < 35%) or observation [6]. Grade 3–4 In these retrospective analyses of heterogenous patients with
cardiopulmonary events occurred in 11% and 5% of patients, and varying degrees of statistical rigour, the heart V5 and V30 flagged
grade 5 in 8% and 0% for the RT and control arms, respectively. in RTOG-0617 were not validated for OS in all datasets [7, 35–37].
This elevated cardiopulmonary event rate associated with post- While other metrics such as MHD did show utility in several
operative RT supports the RTOG-0617 conclusion that heart dose cohorts [38, 39], this was not a consistent trend [7, 8, 40–42],
is an important treatment consideration, although information on aligning with a systematic review that found no consistent
CVRFs and ECDs were also not available for this trial. relationship between any WH metric and OS or cardiac outcomes
RTOG-0617 and LungART provided evidence of significant RIHD [43]. Several studies failed to find a relationship between heart V5
and highlight methodological opportunities to advance the field, and V30 and PRCEs [7, 8, 37], prompting deeper analyses with
such as reporting cardiac baseline factors, a range of heart dose cardiac substructures [44]. Supervised analysis of these distinct
metrics and cardiac-specific outcomes. Systematic reviews indi- regions may be viewed as a systematic and biologically refined
cate that cardiotoxicity is under-represented in NSCLC RT trials approach. Some of the key NSCLC literature on cardiac
[27, 28], suggesting some PRCEs are missed [29]. Commercial trials substructure are summarised in Fig. 2.
may be particularly prone to over-looking cardiotoxicity, as
demonstrated in PACIFIC. This seminal trial integrated immune
checkpoint inhibitors with definitive RT for the first time, OPTIMAL ENDPOINTS
randomising patients with unresectable stage III NSCLC to 1 year In theory, optimal endpoints for RIHD studies would include
of adjuvant durvalumab or placebo [30]. RT dosimetric objectives PRCEs, cardiac mortality, and all-cause mortality, which can help
were not reported, nor were the heart dose metrics, and no capture the competing effects of cancer-specific mortality and
cardiac toxicities were listed in the table of adverse effects [30]. mortality from radiation side effects including RIHD. In practice,
assessing and interpreting these endpoints across studies is
challenging due to heterogenous definitions. Some studies record
RETROSPECTIVE STUDIES all PRCEs that are symptomatic, others use composite cardiology
A total of 42 peer-reviewed papers retrospectively examining RIHD endpoints, while others choose oncological toxicity criteria (e.g.
in NSCLC have been published, excluding those solely focussed on Common Terminology Criteria for Adverse Events), which requires
pericardial endpoints (n = 4) given their non-acute and typically careful interpretation (Fig. 1) [45]. Interpreting grade 3 PRCEs (e.g.
asymptomatic nature (Table 1). Only two studies were published hospitalisation) can be problematic, since the criteria for an
prior to RTOG-0617. Approximately 60% of patients were treated inpatient admission varies greatly between healthcare systems. It
with 3DCRT and the rest with IMRT. The RT planning scan type was is also worth noting that due to their chronic implications, many
evenly distributed between 3D and 4D CT, where reported. Dose- grade 1–2 PRCEs generate the requirement for ongoing for
fractionation details provided were typically insufficient for primary care interactions and subsequent investigations, which

British Journal of Cancer

 4
Table 1. Summary characteristics and findings of the peer-reviewed, retrospective, clinical cardiac dosimetry studies in relation to cardiotoxicity.
Author & year Cohort Stage Average Follow- Type N Baseline Toxicity Planning DVHs MHD Structures Atlas Key study
type EQD2 (Gy) up endpoint constraints findings
(months)
Dess [8] Trial II–III 70 23.0 3DCRT 125 HTN, DM, Primary: G3+ V40 < 100%, Dmean, V5, 11.0 Heart Feng 15% of patients
(97%) “Pre-existing events V65 < 33% V30, V50 had a cardiac
IMRT disease” Others: G2+ event. DVHs
(3%) events, OS positive on MVA.
DVHs positive for
OS on MVA.
Wang [7] (JCO) Trial III 70–90 Gy in NR 3DCRT 112 CAD Primary: V40 < 100%, Dmean, 12.0 Heart Feng 23% patients had
23–45#* Symptomatic LV V40 < 100% V30, V5, LV a cardiac event.
events V30, LV V5 DVHs positive on
Others: OS, MVA.
PCE DVHs negative
for OS on MVA.
Wang [121] Primary: PCE, V40 < 100%, Dmean, V5, Feng for Feng 8%, 7%, 11% of
(R&O) ACS, arrhyth LV V40 < 100% V30, V60 chambers, patients had a
Others: OS LAD PCE, ACS, and
+ automated arrhythmia
pericardium events. DVHs
positive on UVA
(no MVA).
DVHs negative
for OS on MVA.
Borkenhagen RWD I–IV 44–69 Gy in 13.2 NR 76 DM, PVD, Primary: NR Dmean, N/A Atria, Wheatley 7%, 21%, 1% of
[46] 14–34#* “Previous Cardiac Dmax, V30, ventricles, patients had
cardiac events V45 pericardium, atrial arrhythmia
disease” Others: OS coronary effusions, valve
space disease. DVHs
positive on MVA.
DVHs negative
for OS on MVA.
G.M. Walls et al.

Atkins [38] RWD II–III 50–66 Gy in 20.4 3DCRT 748 HTN, Lipids, Primary: None pre- Dmean, V5, 12.3 Heart Feng 10% of patients
(JACC) 1.8–2 Gy #* (76%) DM, VTE, MACE 2008, V30 had MACE. DVHs
IMRT arrhythmia, Others: G3+ V30 < 50%, positive on MVA.
(24%) valve disease, events, OS V45 < 40%, DVHs positive for
PVD, stroke, V60 < 20% OS on MVA.
CAD, MI, HF thereafter
Atkins [73] NR 701 Dmean, Chambers & Feng DVHs positive for
(JAMA Onc) Dmax, Vx in coronaries both MACE and
5 Gy OS.
increments
McKenzie [122] Trial III 60 Gy in 30# 22.9 3DCRT 439 None Primary: OS None Dmean, NR Heart, LAD Feng DVHs positive for
or 74 Gy in (54%) LAD V15 OS.
37#* IMRT
(46%)
Speirs [25] RWD II–III 50–84.9 Gy 14.5 3DCRT 322 None Primary: NR Dmean, 2.6 Heart Wheatley 24% patients had
in (60%) Cardiac Dmax, Vx in (3DCRT) a grade 3 event.
1.5–2.8 Gy IMRT events 5 Gy 1.8 DVHs positive for
#* (40%) Others: OS increments (IMRT) OS on MVA.
Vivekanandan Trial II–III 63–73 Gy in 25.0 3DCRT 78 Pre-existing Primary: OS D100 < 45 Gy, PCAs 10.3 Pericardium, Feng DVHs positive for
[33] 30#* (96%) ECG only Others: ECG D67 < 53 Gy, AVN, 4 OS on MVA.
VMAT change at 6 D33 < 60% chambers, 4 38% patients had
(4%) months chamber ECG changes.
walls DVHs negative
for ECG changes
on MVA.
Vivekanandan RWD I–III 69 38.0 3DCRT 64 “Baseline Primary: OS NR Dmean, Vx 7.6 Heart, LA Feng DVHs positive for
[39] (75%) cardiac in multiple OS on MVA.
VMAT comorbidity” thresholds
(25%)
Yegya-Raman RWD II–IV 60–66 Gy in 47.4 3DCRT 140 CAD, Primary: V40 < 100%, Dmean, V5, 15.8 4x chambers, Feng 29% of patients
[40] 1.8–2 Gy #* (64%) “significant Symptomatic MHD < 26 Gy V30, V50 LAD had an event.
3DCRT arrhythmia, events as per DVHs positive on
+IMRT pericardial Others: OS QUANTEC MVA.
(16%) disease, HF 2010 DVHs negative
IMRT without CAD, for OS on MVA.
(19%) valve disease

British Journal of Cancer

 Table 1. continued
Author & year Cohort Stage Average Follow- Type N Baseline Toxicity Planning DVHs MHD Structures Atlas Key study
type EQD2 (Gy) up endpoint constraints findings
(months)
Yu [123] RWD III 50–72 Gy in 25.5 IMPT 163 HTN, CAD, Primary: NR Dmean, 3 (IMPT) Heart NR 12% of patients
20–36# (21%) DM Cardiac Dmax, V5, / had an event
(IMPT) or IMRT/ events V10, 9 (IMRT) after IMRT/VMAT,
45–66 Gy in VMAT V20, V30, 0% after IMPT.
15–33#* (79%) V40,

British Journal of Cancer

Hotca [81] RWD III 50–80 Gy in 20.0 IMRT 155 DM, DL, HTN, Primary: ECG NR PCAs, Dmin, NR Chambers, Feng 66%, 35% and
1.8–2 Gy #* cardiac Changes Dmean, great vessels 67% of patients
disease (list Dmax had an ECG
included) change that was
arrhythmic,
ischaemic/
pericarditic, or
non-specific.
DVHs positive on
UVA (MVA not
done).
Johnson [124] RWD II–III 60–74 Gy in 16.8 3DCRT 86 None Primary: OS None V5, V30 NR Heart NR DVHs positive for
1.8–2 Gy #* (38%) OS on MVA.
IMRT
(62%)
Han [125] RWD I–III 60–85.5 Gy NR 3DCRT 100 “CVD”, HTN Primary: OS D33 ≤ 66 Gy, Dmax, Vx in NR PA RTOG DVHs positive for
in 2–3.8 Gy D66 ≤ 45 Gy, 5 Gy OS on MVA.
#* D100 ≤ 40 Gy increments
Ma [41] RWD I–III 60–76 Gy in 16.9 3DCRT 141 HTN Primary: OS V30 ≤ 40-60%, Dmean, 5.2 Heart, PA RTOG DVHs positive for
30–38#* (89%) V40 ≤ 30-50% Dmax, Vx in OS on MVA.
IMRT 5 Gy
(11%) increments
Schytte [126] RWD I–III 60–80 Gy in 17.3 3DCRT 250 None Primary: NR Dmean NR LV, bilateral NR 15% patients had
2 Gy #* Cardiac ventricles, a cardiac events.
events (list heart DVHs were
included) negative for both
Others: OS events and OS
on MVA.
Stam [47] RWD II–III 70.1 55.0 IMRT 469 None Primary: OS Dmax ≤40 Gy, V0.5, V1, V2, N/A Heart Feng DVHs positive for
D66 ≤ 50 Gy, V3, V4 OS on MVA.
D33 ≤ 66 Gy and Vx in
5 Gy
increments
G.M. Walls et al.

&
equivalent
uniform
dose
Guberina [35] Trial III 45 Gy in 30# 72.0 3DCRT 155 None Primary: OS NR Dmean, V5 13.8- Heart Feng DVHs negative
± 26 Gy in 17.4 for OS on MVA.
13#*
Sheperd [127] RWD I–III 45–70 Gy in 64.0 3DCRT 284 None Primary: OS V30 ≤ 50% Dmean, 11.2 Heart RTOG DVHs positive for
1.8–2 Gy#* (30%) Dmin, OS on MVA.
IMRT Dmax plus
(70%) Vx in 5 Gy
increments
Tucker [36] RWD III 60–76 Gy in 24.0 3DCRT 468 None Primary: OS NR Dmean, V5, 16.6 Heart NR DVHs negative
1.8–2 Gy#* (41%) V30 for OS on MVA.
IMRT
(49%)
IMPT
(10%)
Lee [42] RWD III 55–70 Gy in 17.6 3DCRT 120 DM, IHD Primary: MI “QUANTEC” Dmean, V5, 12.6 Heart RTOG MI occurred in
1.8–2.75 Gy (42%) (with Others: OS (from 2010) V25, V30, 4%. DVHs
#* IMRT/ definition) V40, V50, negative for MI
VMAT D30 on MVA.
(58%) DVHs positive for
OS on MVA.
5
 6
Table 1. continued
Author & year Cohort Stage Average Follow- Type N Baseline Toxicity Planning DVHs MHD Structures Atlas Key study
type EQD2 (Gy) up endpoint constraints findings
(months)
Lee [128] RWD I–III 50–54 Gy in 36.6 3DCRT 43 DM, IHD Primary: MI Dmean Dmean, V5, 9.4 Heart NR No MI events
1.8–2 Gy #* (70%) (with Others: OS ≤40 Gy, V25, V30, occurred.
IMRT/ definition) V60 ≤ 33%, V40, V50, DVHs negative
VMAT V40 ≤ 80%, D30 for OS on MVA.
(30%) V45 ≤ 60%,
V60 ≤ 33%
Kim [54] RWD I–III 60–64.5 Gy 36.2 3DCRT 321 BMI, DM, AF, Primary: Dmean Dmean, 12.3 Chambers, Feng, Loap 5% of patients
in 30#* (34%) valve disease, Cardiac <45 Gy Dmax, Vx in coronaries, had AF. 2% had
IMRT CAD, CHB, events 5 Gy conduction non-AF events.
(66%) stroke, PVD, Others: OS increments nodes DVHs positive for
CAC, alcohol AF on MVA.
DVHs positive for
OS on MVA.
Banfill [72] RWD I–IV >45 Gy in NR NR 967 HTN, IHD, Primary: V30 < 40%, Dmean, V5, 12.8 Heart UK SABR DVHs positive for
20#* Myocardial, Cardiac death V40 < 30% V30, V50, Consortium OS on MVA.
Pericardial & (from 2015) Guidance
Valve disease,
Arrhythmia
(list included)
Xu [97] Trial II–IV 60–74 Gy in 26.2 IMRT 225 “Pre-existing Primary: V30 < 50%, Dmean, 12.0 Heart, RTOG 25% of patients
2 Gy # * (61%) heart Cardiac V45 < 40%, Dmax, Vx in pericardium, had a cardiac
/ disease” events V60 < 20% 5 Gy chambers event. DVHs not
Protons Others: OS increments positive for
(39%) events on UVA.
(No MVA).
DVHs not
positive for OS
on UVA. (No
MVA).
G.M. Walls et al.

Niska [129] RWD III 43.1–74 Gy 18.0 3DCRT 119 None Primary: OS Dmax <62 Gy, Vx in 5 Gy NR Heart NR DVHs positive for
in 1.8–2 Gy (41%) / Dmean increments OS on MVA.
#* IMRT <26 Gy,
(59%) V30 < 46%,
V40 < 33%
Cho [130] RWD III 60–66 Gy in 45.0 3DCRT 133 Pre-existing Primary: G2+ None Dmean, V5, 8.3 Heart, LV wall NR 32% patients had
2–2.4 Gy #* (83%) heart disease cardiac events mandatory V30, V50 a cardiac event.
/ (list included) DVHs positive for
3DCRT events on MVA.
+IMRT
(3%)
/
IMRT
(14%)
Thor [44] Trial III 60 or 74 Gy 24.0 3DCRT 306 None Primary: OS Heart Dmean, NR Heart, RTOG Model
in 2 Gy #* (52) training V33 < 60 Gy, Dmax, pericardium, combining atria
/ / V66 < 45 Gy, Dmin, bilateral atria, D45, lung
IMRT 131 V100 < 40 Gy MOHX% bilateral Dmean,
(48) validation Dmean, ventricles pericardium
Dmax, and MOH55, and
in 5 Gy ventricles MOH5.
increments
Jang [131] RWD III 50–72 Gy/ 27.5 3DCRT 258 HTN, DM, CV Primary: OS NR Mean, V5, NR WH, RA, LA, NR MVA positive for
25–36# (NR) disease V10, V20, RV, LV LV in patients
/ V30, V40, with cardiac
IMRT V50, and history only
(NR) V60
Yegya-Raman RWD II–IIIC 66–70 Gy/ 39.6 3DCRT 335 HTN, DM, Primary: MHD < 20 Gy, Mean, 8.7 Heart, LV, Feng 10.4% patients
[132] (R&O) 33–35# (6%) / Lipids, MACE V50 < 25% Dmin, LAD had a MACE.
IMRT previous Secondary: Dmax, V5- MVA positive for
(59%)/ atrial Cardiac 70 OS
proton arrhythmia events, OS only, for MHD
(35%) and LAD V15

British Journal of Cancer

 Table 1. continued
Author & year Cohort Stage Average Follow- Type N Baseline Toxicity Planning DVHs MHD Structures Atlas Key study
type EQD2 (Gy) up endpoint constraints findings
(months)
Yegya-Raman RWD II–IV 66 Gy/33# 29.9 3DCRT 187 / 140 Cardiac Primary: MHD < 26 Gy Mean, V5, 8.1 Heart, Feng Ventricle metrics
[112] (ctRO) (11%) / comorbidity, Death V30, V50 MVA positive for
IMRT CAD without DWP and OS.
(36%)/ progression
proton Secondary:

British Journal of Cancer

(57%) OS
Olloni [133] RWD NR 24–33# NR 3DCRT 644 Arrhythmia Primary: OS NR V2 – V70 11.3 WH, RA, LA, NR (auto) MVA positive for
(9%) / and “baseline (2 Gy RV, LV, LMC, OS (2 x PCAs)
IMRT heart intervals) LCX, LAD,
(41%)/ disease” RCA
VMAT
(50%)
McWilliam [61] RWD NR 55 Gy/20# NR 3DCRT 1101 / 89 None Primary: OS NR Dmax NR Heart base N/A Voxel-based
(NR) (SABR) region (not analysis with
/ delineated) positive MVA for
IMRT a dose region
(NR) approximating to
the heart base
McWilliam [66] RWD III 55 Gy/20# NR 3DCRT 978 None Primary: OS NR Dmax NR Heart base N/A Voxel-based
(IJROBP) (NR) region (not analysis with
/ delineated) positive MVA for
IMRT a dose region
(NR) approximating to
the heart base
McWilliam [62] RWD III 55 Gy/20# NR 3DCRT 648 None Primary: OS NR Dmax NR Right atrium N/A Surface dose
(NR) surface(not mapping analysis
/ delineated) with positive
IMRT MVA for a dose
(NR) region
approximating to
the right atrium
McWilliam [64] Trial III 60–74 Gy/ 24 3DCRT 458 None Primary: OS V40 < 100%, V5, V30 NR Heart and NR Voxel-based
30–37# (NR) V45 < 66%, heart base analysis with
/ V60 < 33% region (not positive MVA for
IMRT delineated) a dose region
(NR) approximating to
the heart base
Craddock [65] Trial II–III 60–74 Gy/ NR 3DCRT 172 LVEF Primary: OS V40 ≤ 50% Dmax NR Heart base N/A Voxel-based
G.M. Walls et al.

30–37# (NR) region (not analysis with

/ delineated) positive MVA for
IMRT a dose region
(NR) approximating to
the heart base
Walls [60] RWD I–III 55–84 Gy/ NR 3DCRT 478 HTN, DM, Primary: OS MHD < 26 Gy Dmax NR Heart base Christie 17% patients had
20–42# (29%) / Lipids, Secondary: region Definition a cardiac event.
IMRT previous MI, MI, HF, MVA positive for
(20%)/ previous HF, arrhythmia both endpoints
VMAT previous
(51%) arrhythmia
Walls [57] RWD I–III 55 Gy/20# NR 3DCRT 420 HTN, DM, Primary: AF MHD < 26 Gy Mean, V5, NR Pulmonary Walls 6% patients
(20%)/ Lipids, V55 veins developed AF.
IMRT previous MI, MVA positive
(30%)/ previous HF,
VMAT previous
(50%) arrhythmia
No [134] RWD IIB–IIIC NR 74 3DCRT 233 DM, HTN, Primary: NR Mean, V7, 9.3 WH, LV, LMC, Duane 22% patients had
( < 1%)/ previous CV Myocardial, V15, V27 RCA, LCX, a cardiac event.
IMRT event conductive, LAD, Total Total left
(25%)/ constrictive, left coronary associated with
VMAT valvular arteries cardiac events
(75%) on MVA
7
 G.M. Walls et al.
8

17% experienced

AF and SVT, LMC

can also impact quality of life via time toxicity and polypharmacy

CHB complete heart block, CAC coronary artery calcification, SABR stereotactic ablative radiotherapy, DWP death without progression, VBA voxel-based analysis, MOHX%[Gy] = mean of the hottest x% of the
radiotherapy, HTN hypertension, DM diabetes mellitus, OS overall survival, MVA multivariate analysis, PCE pericardial event, ACS acute coronary syndrome, LAD left anterior descending coronary artery, UVA
univariate analysis, DL dyslipidaemia, VTE venous thromboembolism, PVD peripheral vascular disease, CAD coronary artery disease, MI myocardial infarction, HF heart failure, MACE major adverse cardiac event,
RTOG Radiation Therapy Oncology Group, IMPT intensity-modulated proton therapy, ECG electrocardiogram, AVN atrioventricular node, CVD cardiovascular disease, PA pulmonary artery, LV left ventricle, RV right
ventricle, LA left atrium, RA right atrium, RCA right coronary artery, LMC left main stem coronary artery, LCX left circumflex coronary artery, IHD ischaemic heart disease, BMI body mass index, AF atrial fibrillation,
N number of patients, EQD2 equivalent dose in 2 Gy fractions, NR not reported, DVH dose-volume histogram, MHD mean heart dose, 3DCRT three-dimensional conformal radiotherapy, IMRT intensity-modulated
with VT/asystole,

bradyarrhythmia
associated with

flutter, and the

LCX with atrial
and therefore are likely to negatively impact survivorship.

analyses, PVs
arrhythmia.
Key study
Furthermore, patients that develop low-grade PRCEs may do so

RCA with
findings

On MVA
≥1 G3
because of better health at baseline, and therefore including only
patients with grade 3 events may inadvertently introduce bias.
Importantly, only a minority of studies in NSCLC have found an
association between PRCEs and death [8, 40, 46], suggesting
historically reported PRCEs have not been tightly coupled with
Atlas

Feng

cardiotoxicity-related deaths. Although death events are regarded

as a reliable endpoint in oncology, they are an imperfect endpoint
in RIHD for patients with NSCLC. Investigating the precise
mechanisms by which cardiac dose leads to death is complicated
PVs, nodes
coronaries,
Chambers,
Structures

by the central role played by the heart in the systemic response to

acute medical conditions, leading to difficulty in accurately
discerning the cause of death for patients with lung cancer [26].
The absence of sensitive post-RT cardiac surveillance tests means
occult disease may go undetected. It has been posited that non-
MHD

12.3

event (subclinical) cardiotoxicity is not clinically important until

the onset of an acute medical stressor such as sepsis or
haemorrhage, where it manifests to hamper recovery, increasing
Dmax, Vx in

increments

the risk of death from the intercurrent medical problem [47–49].

Dmean,

Supporting this hypothesis, post-RT cardiac magnetic resonance

DVHs

5 Gy

imaging exhibits cardiac parenchymal changes in asymptomatic

patients [50, 51]. Furthermore, the low rate of fatal PRCEs reported
in the literature does not fully account for the increased death rate
constraints

observed with higher heart doses in NSCLC, positing subclinical

V30 < 50%,
V45 < 40%,
V60 < 20%
None pre-

thereafter
Planning

cardiac insults as a vector for increased mortality. As such, all-

2008,

cause mortality may serve as a surrogate metric for RIHD in large

cohorts where relevant patient, tumour and treatment data are
available. In practice, the magnitude of the cardiac event rate,
Primary: G3

excess death rate and quality of life detriments [52] equally justify
arrhythmia
endpoint

subtypes

the urgent formulation of cardioprotection principles. That cardiac

Toxicity

dose may also behave as a surrogate for other toxicity endpoints

such as pulmonary is also under investigation [53].
valve disease,

CAD, MI, HF
HTN, Lipids,

PVD, stroke,
arrhythmia,
DM, VTE,
Baseline

ANALYTICAL CHALLENGES IN RADIATION CARDIOTOXICITY

STUDIES
There is a wide range of methods available for elucidating the
relationship between radiation dose and toxicity, known as normal
tissue complication probability modelling, and there is no
consensus on the optimal approach. While multivariable regres-
748
N

sion models typically comprise the definitive statistical test of

these studies, there is great variability among the processes by
which substructures, dose metrics and thresholds are selected for
3DCRT
(76%)

(24%)
Type

IMRT

inclusion. Inter- and intra-substructure dose metric multicollinear-

ity complicates this process, in that some proposed high-risk
substructures or critical metrics may actually be a surrogate for an
(months)

adjacent substructure or a similar metric that has greater influence

Follow-

(*EQD2 not reported, so physical prescription provided).

20.4

on the probability of toxicity. For example, a post-RT atrial

up

volume, in Grey, LVEF left ventricular ejection fraction.

fibrillation study reported significant associations for the sinoatrial

node [54], even though this region of specialised myocardium
EQD2 (Gy)

does not have a clear pathophysiological role in this disease. The

1.8–2 Gy#*
50–66 Gy
Average

investigators employed ‘area under the curve’ values to select

candidate dose metrics from a range of almost 150, without any
in

correction for multiple testing [55]. In contrast, a hypothesis-driven

study of the pulmonary veins (PVs) [56, 57] found compelling,
Stage

physiologically sound associations for these structures, and has

II–III

been validated subsequently [58]. These studies found the

relationship of the sinoatrial node to incidence of new atrial
Cohort

fibrillation to be much weaker than that of the PVs, suggesting

RWD
type
continued

that the sinoatrial node is merely a surrogate structure, when

defined as per Loap [59]. While the use of ‘area under the curve’
values to select candidate metrics may maximise the opportunity
Author & year

to find the ground truth, there is the theoretical cost of a higher

Atkins [135]

risk of deriving a false-positive finding [60].

Table 1.

Contemporary computational methods such as voxel-based

analysis may offer an alternative system for identifying critical

British Journal of Cancer

 G.M. Walls et al.
9
Base of heart
Left circumflex coronary artery
Arrhythmia/ischaemia/heart failure
Dmax*–Walls 2024 Atrial flutter
V35 <31%—Atkins 2024

Pulmonary veins
Atrial fibrillation
Left ventricle
V63 <2% (L); V11 <54% (R)—Walls 2024
V5 <15cc—Atkins 2024 Ischaemia
V5*, V30*, mean*—Wang 2017
Supraventricular tachycardia V60 = 0%—Jang 2020
V55 <31%—Atkins 2024
Ischaemia/heart failure
Dmean*—Yegya-Raman 2018
Sinoatrial node
Atrial fibrillation
Dmax 20Gy—Kim 2022 Left anterior descending coronary artery
MACE
V15 <10%—Atkins 2021
Right atrium
Arrhythmia Ischaemia/heart failure
V60*—Wang 2017 Dmean*—Yegya-Raman 2018

Myocardial/constrictive/valve/arrythmia events
V15 <2.5cc–No 2024 (LMC+LAD+LCX)
Right coronary artery
Bradyarrhythmia Ventricular tachycardia/asystole
Right ventricle V50 <0.5cc—Atkins 2024 (LMC)
V25 <1.9cc—Atkins 2024
Ischaemia/heart failure
Dmean*—Yegya-Raman 2018

Fig. 2 Summary of published dosimetric relationships for cardiac endpoints in lung radiotherapy studies. The three-dimensional
reconstruction of autosegmented cardiac substructures from a lung cancer radiotherapy plan, annotated with the literature on post-
radiotherapy cardiac event endpoints by substructure, and proposed dose constraints where available (*no dose threshold reported; Vxx =
percentage volume receiving at least XX Gy; Dmax = maximum dose; Dmean = mean dose).

cardiac regions and their dose thresholds. These high-throughput CVRFs that pre-dispose them to future PRCEs [26]. Potentiating
approaches have been successfully applied to establish the heart this risk, social deprivation in this cohort [67, 68] may lead to
base as a general region of interest [61–63] in real-world datasets, suboptimal health service access and engagement, resulting in
with subsequent retrospective validation in clinical trial cohorts patients being less likely to have their risk factors optimised
[64, 65], and following refinement of the region to a specific [10, 69, 70].
composite of anatomically-defined substructures [66]. In a study of mixed thoracic cancers where patients had cardiac
There is a lack of consensus on how clinical factors that are to stents prior to RT, comorbidity burden appeared to drive survival
be adjusted for in multivariable analysis are selected. Many rather than radiation dose parameters [71]. Similarly, the impact of
studies, especially retrospective studies, base these decisions on the heart base region dose on survival was mediated by the
univariable regression results that meet arbitrary p value thresh- baseline systolic function in a clinical trial cohort [65]. Cardiac
olds, but including known clinically relevant factors may be comorbidity was found to be significantly associated with the risk
stronger. It is important to consider the number of events when of cardiac death after RT at 2 years (21% versus 6%) in another
choosing the number of clinical variables to include in multi- study [72]. Several of the cardiac substructure studies found a
variable models, as over-fitting can occur when an excessive difference in the rate of PRCEs when stratifying by baseline cardiac
amount of variables are included (e.g. 1 variable for every 10 status, including a 2-year MACE rate of 12% versus 3% for patients
events often used as a general principle). with and without cardiac disease [73], and a 4-year incidence of
Lastly, although it is widely acknowledged that inclusion of symptomatic cardiac events of 52% versus 23% where there were
continuous data is superior to dichotomised data in clinical ECDs at baseline compared to without [40]. The relative risk of
models, dichotomisation can be useful in RT toxicity studies, since PRCEs was lower for patients with ECDs, possibly because of the
the output may serve as a directly implementable dose constraint high rate of PRCEs without significant radiation doses. One
for treatment planning. The study design should guide this interpretation of this is that patients with less baseline cardiac
decision, with continuous data perhaps being more important for comorbidity should be prioritised for rigorous cardiac substructure
‘discovery’ studies, while dichotomisation maybe more acceptable avoidance during treatment planning, as the detriment appears to
in studies validating a previously identified metric and threshold. be relatively larger for these patients. However, where such
In summary, methodological research is urgently required for the resources are not limited, the high absolute risk of PRCEs in
most statistically responsible approaches to be identified. patients with a significant cardiac history, means all patients stand
to benefit from these and other cardioprotective strategies.
There is considerable variation between studies regarding the
BASELINE CARDIAC STATUS classification of pre-treatment cardiac comorbidities. Some
The contribution of baseline cardiac disease relative to dosimetric investigators focus on the history of specific cardiac events, such
factors is a current gap in the RIHD literature. The co-morbidity as myocardial infarction (MI) and arrhythmias [72], while others
burden of the general lung cancer population is large, with ~30% focus on pooled ECDs [73]. As the natural history of cardiovascular
of patients having a previous ECD [9] and a further 50% having disease would dictate that ECDs are end-organ consequences of

British Journal of Cancer

 G.M. Walls et al.
10
CVRFs (e.g. hypertension, dyslipidaemia), it is plausible that [89–96], although one group analysed samples from an interven-
patients with risk factors but not yet ECDs are also at heightened tional trial [97]. This study, which compared photon-based IMRT
risk of RIHD. Clinical CV parameters and several CT-based and passive-scattering protons, found that troponin elevations at
correlates have demonstrated utility for selecting a high-risk both baseline and post-RT were associated with PRCEs, irrespec-
group for cardiac assessment and optimisation associated with tive of treatment modality [97]. Overall, that several studies failed
the incidence of PRCEs but not mortality in the NI-HEART cohort to validate signals from animal models reiterates the complex
[74]. Recent European cardio-oncology guidance should improve biology at play in humans, in terms of baseline cardiovascular co-
awareness of baseline risk factors by recommending that all morbidity and partial heart radiation exposures.
patients undergo calculation of a CV risk score [75]. Further
recommendation statements could build on this foundation by
attempting to harmonise how baseline CVRFs and ECDs are CARDIAC SEGMENTATION
recorded. The first peer-reviewed heart atlas was published online in 2010,
Animal data suggests the magnitude of lung irradiation is also aimed at breast cancer radiotherapy planning [31]. This atlas not
relevant when studying RIHD [76], which is logical given these only specified detailed limits for segmentation of the WH, but also
organs are physiologically coupled. However, there is a dearth of for the pericardium and several cardiac substructures. Atlases
clinical data on the impact of baseline pulmonary morbidity. focussing on specific groups of substructures were subsequently
Unfortunately, the inescapable geometric coupling of the heart developed, including atlases for the coronary arteries and LV
and lungs means that any sparing of the heart doses may elevate segments [98, 99], the valves [100], conduction system [59, 101]
the lung dose with standard photon-based treatments [77, 78]. An and pulmonary veins [56]. There is duplication between several of
ongoing photon versus proton study may address this question the atlases, with varying degrees of congruity. For example, the
[79] and functional lung-sparing RT planning techniques may superior vena cava is excluded from the WH contour in the Feng
enable individualised lung dose-sparing [80] in the future if the atlas [31], but included in the Milo atlas [99]. The auricles are
relationship between lung disease and cardiac post-RT outcomes included in the Feng atlas images but not described in the text,
proves significant when evaluated. while Milo recommends including the right but not the left.
Additionally, the Duane [98] and Milo atlases use differing
landmarks to identify the limits of the coronary artery segments.
INVESTIGATIONS FOR DETECTION OF SUBCLINICAL Some of the published atlases have additional weaknesses. For
CARDIAC INJURY example, the inter-operator variability for the valves was high in
In relation to subclinical cardiac dysfunction, ECGs, echocardio- the Socha atlas, and the conduction nodes atlas does not report
grams and other imaging modalities have been tested for the process used to derive the definitions. At present, there are no
assessing asymptomatic toxicity from conventional RT. Among atlases for the great vessels, such as the aorta and pulmonary
the 78 reportable patients from the IDEAL-CRT trial [33], the artery, although it is debated if these vascular structures truly
proportion of patients with a normal ECG dropped from 49% to constitute cardiac substructures per se. Nonetheless, substructure-
23% within 25 months post-RT, with 14% and 12% patients having specific data represent a crucial component of RIHD research,
new ischaemic/pericardial and rhythm changes, respectively. given the wide variety of cardiotoxicity subtypes. For example, in
Furthermore, ECG changes were associated with worse survival. relation to arrhythmia, atrial fibrillation, atrial flutter and sinus
A separate study found that ECG changes were associated with bradycardia originate in the atria, whereas heart block occurs in
reduced survival [81]. This study had higher rates of new the atrioventricular node, and QT prolongation and ventricular
abnormalities (35% ischaemia/pericarditis, 66% arrhythmia) over tachycardia originate in the ventricles.
20 months’ follow-up. This study also identified an association Despite the emerging rationale for considering the cardiac
between higher superior cava minimum radiation dose and new substructures during treatment planning, there are a number of
non-specific ECG abnormalities post-RT. barriers to full implementation, such as added delineation time
Established imaging techniques from cardiology have also been and risk of interobserver variability. The latter issue is secondary to
evaluated for utility in detecting RIHD. Myocardial perfusion the complexity of substructure geometry, limited intracardiac soft
imaging has demonstrated that asymptomatic tissue changes are tissue definition, particularly without contrast enhancement, and
spatially mapped to regions of the highest RT dose distribution motion artefacts. Another unresolved issue in cardiac substructure
[51]. Reductions in cardiac muscle 18FDG uptake on tumour segmentation is whether the cardiac chamber blood pool should
response assessment positron-emission tomography/CT scans be included for the chamber dose calculations. Cardiac sub-
have also been found after RT and were associated with worse structure autosegmentation tools can fortunately overcome many
OS [82]. A recent PET-based radiomic analysis identified quanti- of these barriers [102].
tative imaging features that predict for myocardial SUV changes
following RT [83]. Ventricular strain abnormalities have been
identified with echocardiogram and are more sensitive than TREATMENT PLANNING
ejection fraction in some RT studies [84, 85]. The European cardio- A recent systematic review of published dose constraints noted a
oncology guidelines recommend baseline echocardiography for myriad of options were included during RT treatment plan
patients with an ECD if the heart is within the irradiated volume optimisation, the most common of which were MHD < 26 Gy
[75]. However, high quality data to support these recommenda- (4 studies) and V40 ≤ 30% (3 studies) [103]. Notably, average
tions is lacking, and minimum dose criteria were not outlined. MHD was not different between IMRT (10.9 Gy) and 3DCRT
While there are no clearly validated CT-based markers of RIHD at (10.6 Gy) [103].
present, coronary artery calcification on baseline imaging has Although avoidance of toxicity is important for all organs, the
been tested semi-quantitatively and quantitatively [34, 74, 86, 87] thoracic OARs commonly need to be prioritised for dose-sparing
and was associated with PRCEs in both scenarios. since it is frequently not possible to meet all of the dose
With respect to circulating markers of RIHD, a wide range of constraints. Beyond the long-held belief that many patients with
established (e.g. troponin, NT-proBNP) and novel analytes [88] intrathoracic cancers do not survive long enough to experience
have been investigated, without a clearly superior option being RIHD, the heart has been given a low priority historically for
identified (Supplementary Table 2). However, studies have several other practical reasons. Firstly, cardiotoxicity rates were
generally consisted of small, mixed tumour cohorts, with a range lower than oesophagitis or pneumonitis. The low-quality evidence
of dose-fractionations and varied timepoints for blood sampling base for the putative WH dose constraints also failed to convince

British Journal of Cancer

 G.M. Walls et al.
11
some clinicians that negative prognostic correlates of heart dose chemotherapy [109]. However, the independent risk of cardio-
do not merely reflect unfavourable disease (e.g. node-positive) toxicity from cytotoxic therapies is well established [110].
[104]. Lastly, there have been no globally agreed-upon guidelines Unfortunately, many lung radiotherapy studies do not include
for heart delineation. Some definitions include significant portions this factor in multivariable analyses, possibly because the large
of the proximal great vessels so that their encapsulating academic centres involved have very low rates of radiotherapy
pericardium is included (Supplemental Table 1), despite the fact without chemotherapy, due to the select patient population
that symptomatic pericardial problems are infrequent in the era of treated. In those studies where chemotherapy was adjusted for,
conformal RT. As well as offering false reassurance that the heart none reported a significant relationship with cardiac endpoints in
dose is low for many cases, the dose summary values from this multivariable analysis [34, 57, 72, 111]. This trend may be due to
version of the WH structure have less relevance for the critical the fact that some of the higher risk systemic therapies such as
cardiac events. For as long as cardiac doses are assessed on a WH anthracyclines are not typically employed in the management of
basis in clinical practice, it may be prudent that critical cardiac lung cancer. Interestingly, one group investigated if there was a
substructures are reviewed on all CT slices in order to identify mal- difference between cisplatin and carboplatin delivered with RT
placed hot spots, e.g. in the LAD. and found the former to have a significant relationship with
For dosimetrists to integrate cardiac substructures in routine cardiotoxicity [112]. Only one published study has examined if
clinical workflows, several adaptations of the current RT pathway durvalumab, the adjuvant immunotherapy that has been standard
for NSCLC are needed. Firstly, a high quality simulation, including of care after lung cancer RT since 2018, is associated with
respiratory management and use of contrast agents where cardiotoxicity, and the authors found no evidence of a relationship
appropriate, is required for the cardiac landmarks to be [113]. Given the potential acceleration of atherosclerosis that has
identifiable. Secondly, more streamlined delineation approaches been reported with such immune checkpoint-based immunother-
are needed as this is currently time-consuming, particularly during apy [114], further study of this factor in future studies continues to
an operator’s learning phase. Thirdly, putative dose constraints for be warranted, and novel drugs being trialled as radiotherapy–drug
the optimiser software to attain will be required if plans are to be combinations [115] should also be carefully examined.
cardiac-optimised, because the ALARA principle may not be
maximally effective for such spatially clustered structures. Because
of the small size of the many of these structures relative to the DISCUSSION
magnitude of respiratory motion, motion envelopes and planning The heart is arguably one of the most heterogenous OARs,
risk volumes will be important. comprised of spatially distinct tissue types with differing radio-
Several groups have explored the feasibility of incorporating biological characteristics, plus functional inter-dependence and
cardiac substructures in silico. One team re-optimised plans for geometric instability. Cardiotoxicity is a common adverse effect of
cardiac substructures and achieved significantly lower values for definitive RT in patients with lung cancer that results in physical
30 of 32 cardiac substructure dose-volume parameters, using a deconditioning, hospitalisation, and non-cancer death during
contemporary version of IMRT known as volumetric modulated arc survivorship. RIHD can be distressing for patients with lung
therapy (VMAT) [105]. While the number of arcs (and treatment cancer, who commonly concurrently deal with symptoms of lung
time) were increased compared to planning without substructure disease, treatment sequelae and the lung cancer itself. For
optimisation, the cardiac-optimised plans did not result in greater clinicians, RIHD continues to represents a major barrier to
doses to other thoracic OARs. Similarly, another arc therapy study treatment intensification as evidenced by the negative dose
demonstrated that introducing a left atrium dose constraint was escalation and acceleration studies.
feasible and could lead to multiple other dosimetric benefits Thirty years after Emami and colleagues omitted lung cancer
including lower lung doses [106]. Additionally, similar findings data in their landmark papers on cardiac dose-volume character-
were also observed in an MRI-based mixed thoracic cancers istics, lung cancer continues to evade rigorous cardiac radio-
planning study [107]. The authors of this paper also reported the biological interrogation. Ultimately, a collaborative research effort
cardiac substructure dose objectives utilised for optimisation. between translational radiobiologists, cardiovascular academics
Another priority for radiation physicists is motion management and radiation oncologists will be required in order to progress the
for the cardiac substructures, given their cyclical shape changes, field, so that the clinical goals of prevention, early detection and
which are irregular in space due to motion from the lungs. As prompt treatment of RIHD can be met. Issues raised in Emami’s
modern RT planning often accounts for tumour motion, a influential paper still hold true for RIHD:
planning volume at risk (PRV) margin for the cardiac substructures
may be warranted. PRV margins of 5.8 mm and 4.8 mm were “Comparative overview of radiation dose tolerance of normal
recommended for compensation of WH motion in the lateral and tissue is a major task… The diversity of organs, endless variations
cranio-caudal axes respectively [108]. However, independent in any combination of radiotherapeutic parameters such as
displacement of the individual cardiac substructures is likely to fractionation, volume, overall time, etc, physiological status of the
be under-estimated by WH margins, as virtually any plane through organs prior to the commencement of radiation… are some of
the heart contains multiple substructures exhibiting non- the factors that makes this task an enormous undertaking.”
synchronised and non-isotropic deformation. Furthermore, sub-
structure motion cannot be assumed to be oscillatory after Despite a century of pre-clinical research, there is a severe lack
interactions from concurrent respiratory motion are considered. It of RIHD experiments in co-morbidity animal models [116]. While
would be important that adoption of such technical approaches many features of cardiac radiopathology have been successfully
be founded through robust evidence in order to justify the described in isolation with pre-clinical methods [117], the multi-
additional resource required, and at present there are no such tude of inter-connected dose- and time-dependent and spatially-
supportive data, possibly becasue the pragmatic trials required are specific pathophysiologies underpinning RHD have not been
difficult to design and fund. probed with contemporary biomedical research pipelines. Further-
more, although multiple studies have shown a cardiac radio-
protective effect for several drugs, none of these have been tested
CONCOMITANT SYSTEMIC ANTICANCER THERAPIES in a clinical trial. Statin therapies have the greatest amount of
Another clinically relevant factor is the chemotherapy delivered evidence in this area [118–120] but as most patients with lung
alongside RT, which is serves to radiosensitise lung tumours, as cancer meet the criteria for a statin on cardiovascular grounds,
exemplified by increased survival rates compared to RT without conducting a randomised controlled trial is not warranted or

British Journal of Cancer

 G.M. Walls et al.
12
Clinical priorities Research priorities

x Relative weight of CVRFs and ECDs

x Record baseline cardiac status AT-RISK POPULATION x RT planning scan features
x Optimise via Primary Care/Cardiology
x Pharmacologic cardioprotection
x Personalised radiotherapy consent

x Dosimetric effects of sparing substructures

x (Auto)segment cardiac substructures
x Principles of substructure prioritisation
x Review cardiac dose slice-by-slice TREATMENT PLANNING
x Impact of PRVs and excluding blood
x Inform Primary Care on relevant doses
x Consideration of concomitant SACT

x Regular ECG monitoring EVENT DIAGNOSIS x Optimal modalities for subclinical RIHD
x Consider cardiac causes of symptoms x Relative contribution of radiation / baseline

Fig. 3 Suggested priorities at each stage of the patient journey for both clinicians and researchers. (CVRF = cardiovascular risk factor; ECD
= established cardiac disease; ECG = electrocardiogram; RIHD = radiation-induced heart disease; PRV = planning organ at risk volume).

feasible. Rather, radiation oncologists are well positioned to take 4. Nygård L, Vogelius IR, Fischer BM, Kjær A, Langer SW, Aznar MC, et al. A Com-
stewardship of such primary prevention therapy when assessing peting Risk Model of First Failure Site after Definitive Chemoradiation Therapy
new NSCLC cases. It is also prudent that patients are fully informed for Locally Advanced Non–Small Cell Lung Cancer. J Thorac Oncol.
of the risk of RIHD in meaningful terms, so that help is sought 2018;13:559–67.
5. Bradley JD, Hu C, Komaki RR, Masters GA, Blumenschein GR, Schild SE, et al.
promptly should they be affected. Active monitoring of select
Long-term results of NRG oncology RTOG 0617: Standard-versus high-dose
cases following stratification by baseline risk and any added RT- chemoradiotherapy with or without cetuximab for unresectable stage III non-
associated risk, or pre-treatment cardiac optimisation, for which small-cell lung cancer. J Clin Oncol. 2020;38:706–14.
the feasibility is being evaluated in the CARMA study 6. le Pechoux C, Pourel N, Barlesi F, Lerouge D, Antoni D, Lamezec B, et al. Post-
(NCT05403736), are expected to significantly improve the cardiac operative radiotherapy versus no postoperative radiotherapy in patients with
safety of lung cancer RT. completely resected non-small-cell lung cancer and proven mediastinal N2
In the future, it is possible that treatment planning will involve involvement (Lung ART): an open-label, randomised, phase 3 trial. Lancet Oncol.
enhanced individualised treatment planning goals in order to 2022;23:104–14.
capitalise on all available baseline risk information, in terms of the 7. Wang K, Eblan MJ, Deal AM, Lipner M, Zagar TM, Wang Y, et al. Cardiac toxicity
after radiotherapy for stage III non-small-cell lung cancer: pooled analysis of
tumour location and pre-existing cardiac health. For example, a
dose-escalation trials delivering 70 to 90 Gy. J Clin Oncol. 2017;35:1387–94.
patient with excellent pulmonary function but severe coronary 8. Dess RT, Sun Y, Matuszak MM, Sun G, Soni PD, Bazzi L, et al. Cardiac events after
artery disease and a medial left lower lobe tumour might be radiation therapy: combined analysis of prospective multicenter trials for locally
permitted to have lung doses slightly above the local constraint to advanced non-small-cell lung cancer. J Clin Oncol. 2017;35:1395–402.
enable the LAD to be spared. Adjustment for cytotoxic, immune 9. García-Pardo M, Chang A, Schmid S, Dong M, Brown MC, Christiani D et al.
checkpoint inhibitors and emerging novel concomitant systemic Respiratory and cardiometabolic comorbidities and Stage I-III non-small cell
therapies delivered with RT is also vital. lung cancer (NSCLC) survival: A pooled analysis from the International Lung
Suggested current priorities for the clinicians and researchers Cancer Consortium (ILCCO). Journal of Thoracic Oncology 2022; 0. https://
are summarised in Fig. 3. The RTOG 1308 randomised controlled doi.org/10.1016/j.jtho.2022.10.020.
10. Kimenai DM, Pirondini L, Gregson J, Prieto D, Pocock SJ, Perel P, et al. Socio-
trial (NCT01993810) comparing photons versus photons in stage III
economic Deprivation: An Important, Largely Unrecognized Risk Factor in Pri-
NSCLC has included RIHD within the composite primary endpoint mary Prevention of Cardiovascular Disease. Circulation. 2022;146:240–8.
and is likely to provide high-quality data on this topic when it 11. Rössle R. Specific roentgen injury of the heart muscle. Strahlentherapie.
reports in 2026 An international prospective registry of patients 1924;XVIII:812–28.
with lung cancer undergoing definitive RT would address many of 12. Davis KS. Intrathoracic changes following x-ray treatment: a clinical and
the standardisation issues. Preparatory work will be needed to experimental study. Radiology. 1924;3:301.
elicit the most statistically responsible approach to such analyses, 13. Cohn KE, Stewart JR, Pajardo LF, Hancock EW. Heart disease following radiation.
in terms of multiple testing and multicollinearity. Ongoing Medicine. 1967;46:281–98.
prospective trials particularly CLARITY (NCT04305613) is likely to 14. PORT Meta-analysis Trialists Group. Postoperative radiotherapy in non-small-cell
lung cancer: Systematic review and meta-analysis of individual patient data
generate high quality data on how established and experimental
from nine randomised controlled trials. Lancet. 1998;352:257–63.
blood-, ECG- and imaging-based parameters could be leveraged 15. Emami B, Lyman J, Brown A, Cola L, Goitein M, Munzenrider JE, et al. Tolerance
to identify at-risk patients, and how to actively monitor for toxicity. of normal tissue to therapeutic irradiation. Int J Radiat Oncol Biol Phys.
Robust prospective clinical and biomarker data will serve to inform 1991;21:109–22.
the design of cardioprotective clinical trials for patients with 16. Burman C, Kutcher GJ, Emami B, Goitein M. Fitting of normal tissue tolerance
NSCLC. data to an analytic function. Int J Radiat Oncol Biol Phys. 1991;21:123–35.
17. Schultz-Hector S, Sund M, Thames HD. Fractionation response and repair
kinetics of radiation-induced heart failure in the rat. Radiother Oncol.
REFERENCES 1992;23:33–40.
1. Patz EF, Pinsky P, Gatsonis C, Sicks JRD, Kramer BS, Tammemägi MC, et al. 18. Lauk S, Rüth S, Trott KR. The effects of dose-fractionation on radiation-induced
Overdiagnosis in low-dose computed tomography screening for lung cancer. heart disease in rats. Radiother Oncol. 1987;8:363–7.
JAMA Intern Med. 2014;174:269–74. 19. Gillette EL, McChesney SL, Hoopes PJ. Isoeffect curves for radiation-induced
2. Siegel RL, Giaquinto AN, Jemal A Cancer statistics, 2024. CA Cancer J Clin 2024; cardiomyopathy in the dog. Int J Radiat Oncol Biol Phys. 1985;11:2091–7.
74. https://doi.org/10.3322/CAAC.21820. 20. Douglas BG, Fowler JF. The effect of multiple small doses of X rays on skin
3. Cancer Research UK. A cancer of substantial unmet need: why we’re making reactions in the mouse and a basic interpretation. Radiat Res. 1976;66:401–26.
lung cancer a priority. 2015. https://news.cancerresearchuk.org/2015/07/20/a- 21. Gagliardi G, Lax I, Ottolenghi A, Rutqvist LE. Long-term cardiac mortality after
cancer-of-substantial-unmet-need-why-were-making-lung-cancer-a-priority/ radiotherapy of breast cancer-application of the relative seriality model. Br J
(accessed 3 Dec2022). Radio. 1996;69:839–46.

British Journal of Cancer

 G.M. Walls et al.
13
22. Gagliardi G, Constine LS, Moiseenko V, Correa C, Pierce LJ, Allen AM et al. Cancer? A Systematic Review of the Literature. Int J Radiat Oncol Biol Phys.
Radiation dose-volume effects in the heart. Int J Radiat Oncol Biol Phys 2010; 76. 2019;104:582–9.
https://doi.org/10.1016/J.IJROBP.2009.04.093. 44. Thor M, Deasy JO, Hu C, Gore E, Bar-Ad V, Robinson C, et al. Modeling the impact
23. Chun SG, Hu C, Choy H, Komaki RU, Timmerman RD, Schild SE, et al. Impact of of cardiopulmonary irradiation on overall survival in NRG oncology trial RTOG
Intensity-Modulated Radiation Therapy Technique for Locally Advanced Non- 0617. Clin Cancer Res. 2020;26:4643–50.
Small-Cell Lung Cancer: A Secondary Analysis of the NRG Oncology RTOG 0617 45. Herrmann J, Lenihan D, Armenian S, Barac A, Blaes A, Cardinale D, et al. Defining
Randomized Clinical Trial. J Clin Oncol. 2017;35:56–62. cardiovascular toxicities of cancer therapies: an International Cardio-Oncology
24. Bradley JD, Paulus R, Komaki R, Masters G, Blumenschein G, Schild S, et al. Society (IC-OS) consensus statement. Eur Heart J. 2022;43:280–99.
Standard-dose versus high-dose conformal radiotherapy with concurrent and 46. Borkenhagen JF, Bergom C, Rapp CT, Klawikowski SJ, Rein LE, Gore EM. Dosi-
consolidation carboplatin plus paclitaxel with or without cetuximab for patients metric predictors of cardiotoxicity in thoracic radiotherapy for lung cancer. Clin
with stage IIIA or IIIB non-small-cell lung cancer (RTOG 0617): A randomised, Lung Cancer. 2019;20:435–41.
two-by-two factorial phase 3 study. Lancet Oncol. 2015;16:187–99. 47. Stam B, van der Bijl E, van Diessen J, Rossi MMG, Tijhuis A, Belderbos JSA, et al.
25. Speirs CK, DeWees TA, Rehman S, Molotievschi A, Velez MA, Mullen D, et al. Heart dose associated with overall survival in locally advanced NSCLC patients
Heart dose is an independent dosimetric predictor of overall survival in locally treated with hypofractionated chemoradiotherapy. Radiother Oncol. 2017;125:62–5.
advanced non-small cell lung cancer. J Thorac Oncol. 2017;12:293–301. 48. Wong OY, Yau V, Kang J, Glick D, Lindsay P, Le LW, et al. Survival Impact of
26. Sun F, Franks K, Murray L, Lilley J, Wheller B, Banfill K, et al. Cardiovascular Cardiac Dose Following Lung Stereotactic Body Radiotherapy. Clin Lung Cancer.
mortality and morbidity following radical radiotherapy for lung cancer: Is car- 2018;19:e241–e246.
diovascular death under-reported? Lung Cancer. 2020;146:1–5. 49. Andratschke N, Maurer J, Molls M, Trott KR. Late radiation-induced heart disease
27. Or M, Liu B, Lam J, Vinod S, Xuan W, Yeghiaian-Alvandi R et al. A systematic after radiotherapy. Clin importance, radiobiological mechanisms Strateg Prev
review and meta-analysis of treatment-related toxicities of curative and pallia- Radiother Oncol. 2011;100:160–6.
tive radiation therapy in non-small cell lung cancer. Sci Rep 2021; 11. https:// 50. Umezawa R, Ota H, Takanami K, Ichinose A, Matsushita H, Saito H, et al. MRI
doi.org/10.1038/S41598-021-85131-7. findings of radiation-induced myocardial damage in patients with oesophageal
28. Prasad RN, Miller ED, Addison D, Bazan JG. Lack of Cardiotoxicity Endpoints in cancer. Clin Radio. 2014;69:1273–9.
Prospective Trials Involving Chest Radiation Therapy: A Review of Registered, 51. Gayed IW, Liu HH, Wei X, Liao Z, Yusuf SW, Chang JY, et al. Patterns of cardiac
Latter-Phase Studies. Front Oncol. 2022;12:328. perfusion abnormalities after chemoradiotherapy in patients with lung cancer. J
29. Prasad RN, Mcintyre M, Guha A, Carter RR, Yildiz VO, Paskett E, et al. Cardio- Thorac Oncol. 2009;4:179–84.
vascular Event Reporting in Modern Cancer Radiation Therapy Trials. Adv Radiat 52. Hui R, Özgüroğlu M, Villegas A, Daniel D, Vicente D, Murakami S, et al. Patient-
Oncol. 2022;7:100888. reported outcomes with durvalumab after chemoradiotherapy in stage III,
30. Spigel DR, Faivre-Finn C, Gray JE, Vicente D, Planchard D, Paz-Ares L, et al. Five- unresectable non-small-cell lung cancer (PACIFIC): a randomised, controlled,
Year Survival Outcomes From the PACIFIC Trial: Durvalumab After Chemor- phase 3 study. Lancet Oncol. 2019;20:1670–80.
adiotherapy in Stage III Non-Small-Cell Lung Cancer. J Clin Oncol. 53. Huang EX, Hope AJ, Lindsay PE, Trovo M, El Naqa I, Deasy JO, et al. Heart
2022;40:1301–11. irradiation as a risk factor for radiation pneumonitis. Acta Oncol. 2011;50:51–60.
31. Feng M, Moran JM, Koelling T, Chughtai A, Chan JL, Freedman L, et al. Devel- 54. Kim KH, Oh J, Yang G, Lee J, Kim J, Gwak S et al. Association of sinoatrial node
opment and validation of a heart atlas to study cardiac exposure to radiation radiation dose with atrial fibrillation and mortality in patients with lung cancer.
following treatment for breast cancer. Int J Radiat Oncol Biol Phys. JAMA Oncol 2022. https://doi.org/10.1001/JAMAONCOL.2022.4202.
2011;79:10–18. 55. Walls GM, Hanna GG. Sinoatrial Node Radiation Dose and Atrial Fibrillation in
32. Radiation Thoracic Oncology Group. Atlases for Organs at Risk (OARs) in Thor- Patients With Lung Cancer. JAMA Oncol. 2023;9:573–4.
acic Radiation Therapy. 2011. https://www.eviq.org.au/getmedia/a4c012a8- 56. Walls GM, McCann C, Ball P, Atkins KM, Mak RH, Bedair A et al. A pulmonary vein
d6a7-4d87-93f7-d3f465b49889/RTOG-heart-contouring-atlas.pdf.aspx?ext=.pdf atlas for radiotherapy planning. Radiother Oncol 2023; 184. https://doi.org/
(accessed 12 May2024). 10.1016/J.RADONC.2023.109680.
33. Vivekanandan S, Landau DB, Counsell N, Warren DR, Khwanda A, Rosen SD, et al. 57. Walls GM, McCann C, O’Connor J, O’Sullivan A, I Johnston D, McAleese J, et al.
The impact of cardiac radiation dosimetry on survival after radiation therapy for Pulmonary vein dose and risk of atrial fibrillation in patients with non-small cell
non-small cell lung cancer. Int J Radiat Oncol Biol Phys. 2017;99:51–60. lung cancer following definitive radiotherapy: an NI-HEART analysis. Radiother
34. No HJ, Guo FB, Park NJ-I, Kastelowitz N, Rhee J-W, Clark DE et al. Predicting adverse Oncol. 2024;192:110085.
cardiac events after radiotherapy for locally advanced non–small cell lung cancer. 58. Atkins KA ZSKCGCHJGJBMSKKDZPNANAMR. Association of cardiac substructure
JACC CardioOncol 2023. https://doi.org/10.1016/j.jaccao.2023.08.007. radiation dose with tachyarrhythmia and bradyarrhythmia after lung cancer
35. Guberina M, Eberhardt W, Stuschke M, Gauler T, Heinzelmann F, Cheufou D, radiotherapy. JACC CardioOncol 2024; in press.
et al. Heart dose exposure as prognostic marker after radiotherapy for resect- 59. Loap P, Servois V, Dhonneur G, Kirov K, Fourquet A, Kirova Y A Radiation
able stage IIIA/B non-small-cell lung cancer: secondary analysis of a randomized Therapy Contouring Atlas for Cardiac Conduction Node Delineation. Pract
trial. Ann Oncol. 2017;28:1084–9. Radiat Oncol 2021; 11. https://doi.org/10.1016/J.PRRO.2021.02.002.
36. Tucker SL, Liu A, Gomez D, Tang LL, Allen P, Yang J, et al. Impact of heart and 60. Walls G, McWilliam A. Association of Cardiac Substructure Radiation Dose With
lung dose on early survival in patients with non-small cell lung cancer treated Arrhythmia: Time to Move Away From Mean Dose. JACC CardioOncol. 2024;6:557–9.
with chemoradiation. Radiother Oncol. 2016;119:495–500. 61. McWilliam A, Kennedy J, Hodgson C, Vasquez Osorio E, Faivre-Finn C, van Herk
37. Xue J, Han C, Jackson A, Hu C, Yao H, Wang W, et al. Doses of radiation to the M. Radiation dose to heart base linked with poorer survival in lung cancer
pericardium, instead of heart, are significant for survival in patients with non- patients. Eur J Cancer. 2017;85:106–13.
small cell lung cancer. Radiother Oncol. 2019;133:213–9. 62. McWilliam A, Dootson C, Graham L, Banfill K, Abravan A, van Herk M. Dose
38. Atkins KM, Rawal B, Chaunzwa TL, Lamba N, Bitterman DS, Williams CL, et al. surface maps of the heart can identify regions associated with worse survival for
Cardiac radiation dose, cardiac disease, and mortality in patients with lung lung cancer patients treated with radiotherapy. Phys Imaging Radiat Oncol.
cancer. J Am Coll Cardiol. 2019;73:2976–87. 2020;15:46–51.
39. Vivekanandan S, Fenwick JD, Counsell N, Panakis N, Stuart R, Higgins GS, et al. 63. Liu X, Fatyga M, Schild SE, Li J. Detecting spatial susceptibility to cardiac toxicity
Associations between cardiac irradiation and survival in patients with non-small of radiation therapy for lung cancer. IISE Trans Health Syst Eng. 2020;10:243–50.
cell lung cancer: Validation and new discoveries in an independent dataset. 64. McWilliam A, Abravan A, Banfill K, Faivre-Finn C, van Herk M. Demystifying the
Radiother Oncol. 2021;165:119–25. results of rtog 0617: identification of dose sensitive cardiac subregions asso-
40. Yegya-Raman N, Wang K, Kim S, Reyhan M, Deek MP, Sayan M, et al. Dosimetric ciated with overall survival. J Thorac Oncol. 2023;18:599–607.
predictors of symptomatic cardiac events after conventional-dose chemor- 65. Craddock M, Nestle U, Koenig J, Schimek-Jasch T, Kremp S, Lenz S et al. Cardiac
adiation therapy for inoperable NSCLC. J Thorac Oncol. 2018;13:1508–18. function modifies the impact of heart base dose on survival: a voxel-wise
41. Ma JT, Sun L, Sun X, Xiong ZC, Liu Y, Zhang SL, et al. Is pulmonary artery a dose- analysis of patients with lung cancer from the pet-plan trial. J Thorac Oncol
limiting organ at risk in non-small cell lung cancer patients treated with defi- 2022. https://doi.org/10.1016/J.JTHO.2022.09.004.
nitive radiotherapy. Radiat Oncol. 2017;12:1–10. 66. McWilliam A, Khalifa J, Vasquez Osorio E, Banfill K, Abravan A, Faivre-Finn C, et al.
42. Lee CC, Zheng H, Soon YY, Foo LL, Koh WY, Leong CN, et al. Association novel methodology to investigate the effect of radiation dose to heart sub-
between radiation heart dosimetric parameters, myocardial infarct and overall structures on overall survival. Int J Radiat Oncol Biol Phys. 2020;108:1073–81.
survival in stage 3 non-small cell lung cancer treated with definitive thoracic 67. Cheng E, Soulos PR, Irwin ML, Cespedes Feliciano EM, Presley CJ, Fuchs CS, et al.
radiotherapy. Lung Cancer. 2018;120:54–59. Neighborhood and Individual Socioeconomic Disadvantage and Survival
43. Zhang TW, Snir J, Boldt RG, Rodrigues GB, Louie AV, Gaede S, et al. Is the Among Patients With Nonmetastatic Common Cancers. JAMA Netw Open.
Importance of Heart Dose Overstated in the Treatment of Non-Small Cell Lung 2021;4:e2139593–e2139593.

British Journal of Cancer

 G.M. Walls et al.
14
68. Li X, Sundquist J, Zöller B, Sundquist K. Neighborhood Deprivation and Lung 90. Kozak KR, Hong TS, Sluss PM, Lewandrowski EL, Aleryani SL, MacDonald SM,
Cancer Incidence and Mortality: A Multilevel Analysis from Sweden. J Thorac et al. Cardiac blood biomarkers in patients receiving thoracic (chemo)radiation.
Oncol. 2015;10:256–63. Lung Cancer. 2008;62:351–5.
69. Brown SH, Griffith ML, Kripalani S, Horst SN. Association of Health Literacy and 91. Gomez DR, Yusuf SW, Munsell MF, Welsh JW, Liao Z, Lin SH, et al. Prospective
Area Deprivation With Initiation and Completion of Telehealth Visits in Adult exploratory analysis of cardiac biomarkers and electrocardiogram abnormalities
Medicine Clinics Across a Large Health Care System. JAMA Netw Open. in patients receiving thoracic radiation therapy with high-dose heart exposure. J
2022;5:e2223571–e2223571. Thorac Oncol. 2014;9:1554–60.
70. Kurani SS, Lampman MA, Funni SA, Giblon RE, Inselman JW, Shah ND, et al. Asso- 92. Serrano NA, Mikkelsen R, Canada J, Mezzaroma E, Weiss E, Abbate A. Biomarkers
ciation Between Area-Level Socioeconomic Deprivation and Diabetes Care Quality in of cardiac injury in patients undergoing thoracic radiation therapy. Int J Cardiol.
US Primary Care Practices. JAMA Netw Open. 2021;4:e2138438–e2138438. 2016;223:507–9.
71. Sio TT, Liang JJ, Chang K, Novotny PJ, Prasad A, Miller RC. Outcomes After 93. Ekici K, Cakal B, Baydar O, Mayadagli A, Cakal SD, Karaca O, et al. Measurement
Curative Thoracic Radiotherapy in Patients With Coronary Artery Disease and of strain imaging, troponin-I and brain natriuretic peptide levels in radiotherapy-
Existing Cardiac Stents. Am J Clin Oncol. 2016;39:549–55. induced cardiac injury. UHOD - Uluslararasi Hematol-Onkol Derg. 2016;26:75–82.
72. Banfill K, Abravan A, van Herk M, Sun F, Franks K, McWilliam A et al. Heart dose 94. Hawkins PG, Sun Y, Dess RT, Jackson WC, Sun G, Bi N et al. Circulating microRNAs as
and cardiac comorbidities influence death with a cardiac cause following biomarkers of radiation-induced cardiac toxicity in non-small-cell lung cancer. J
hypofractionated radiotherapy for lung cancer. Front Oncol 2022; 12. https:// Cancer Res Clin Oncol 2019; 145. https://doi.org/10.1007/S00432-019-02903-5.
doi.org/10.3389/FONC.2022.1007577. 95. Kuo AH, Ancukiewicz M, Kozak KR, Yock TI, Padera TP Cardiac and inflammatory
73. Atkins KM, Chaunzwa TL, Lamba N, Bitterman DS, Rawal B, Bredfeldt J, et al. biomarkers do not correlate with volume of heart or lung receiving radiation.
Association of left anterior descending coronary artery radiation dose with Radiat Oncol 2015; 10. https://doi.org/10.1186/S13014-014-0324-1.
major adverse cardiac events and mortality in patients with non-small cell lung 96. Canada JM, Thomas GK, Trankle CR, Carbone S, Billingsley H, Van Tassell BW
cancer. JAMA Oncol. 2021;7:206–19. et al. Increased C-reactive protein is associated with the severity of thoracic
74. Walls GM, Hill N, McMahon M, Kearney BÓ, McCann C, McKavanagh P et al. radiotherapy-induced cardiomyopathy. Cardiooncology 2020; 6. https://doi.org/
Baseline cardiac parameters as biomarkers of radiation cardiotoxicity in lung 10.1186/S40959-020-0058-1.
cancer: an NI-HEART analysis. JACC CardioOncol 2024; in press. 97. Xu T, Meng QH, Gilchrist SC, Lin SH, Lin R, Xu T, et al. Clinical investigation
75. Lyon A, López-Fernández T, Couch LS, Asteggiano R, Aznar MC, Bergler-Klein J, assessment of prognostic value of high-sensitivity cardiac troponin t for early
et al. 2022 ESC Guidelines on cardio-oncology developed in collaboration with prediction of chemoradiation therapy-induced cardiotoxicity in patients with
the European Hematology Association (EHA), the European Society for Ther- non-small cell lung cancer: a secondary analysis of a prospective randomized
apeutic Radiology and Oncology (ESTRO) and the International Cardio-Oncology trial. Int J Radiat Oncol Biol Phys. 2021;111:2021.
Society (IC-OS). Eur Heart J. 2022;43:4229–361. 98. Duane F, Aznar MC, Bartlett F, Cutter DJ, Darby SC, Jagsi R, et al. A cardiac
76. Wiedemann J, Coppes RP, van Luijk P Radiation-induced cardiac side-effects: contouring atlas for radiotherapy. Radiother Oncol. 2017;122:416.
The lung as target for interacting damage and intervention. Front Oncol 2022; 99. Milo MLH, Offersen BV, Bechmann T, Diederichsen ACP, Hansen CR, Holtved E,
12. https://doi.org/10.3389/FONC.2022.931023. et al. Delineation of whole heart and substructures in thoracic radiation therapy:
77. Storey CL, Hanna GG, Greystoke A. Practical implications to contemplate when National guidelines and contouring atlas by the Danish Multidisciplinary Cancer
considering radical therapy for stage III non-small-cell lung cancer. Br J Cancer. Groups. Radiother Oncol. 2020;150:121–7.
2020;123:28–35. 100. Socha J, Rygielska A, Uziębło-Życzkowska B, Chałubińska-Fendler J, Jurek A,
78. Conibear J. Rationale for concurrent chemoradiotherapy for patients with stage Maciorowska M, et al. Contouring cardiac substructures on average intensity
III non-small-cell lung cancer. Br J Cancer. 2020;123:10–17. projection 4D-CT for lung cancer radiotherapy: A proposal of a heart valve
79. Baumann BC, Hallahan DE, Michalski JM, Perez CA, Metz JM. Concurrent chemo- contouring atlas. Radiother Oncol. 2022;167:261–8.
radiotherapy with proton therapy: reduced toxicity with comparable oncologi- 101. Schottstaedt AM, Paulson ES, Rubenstein JC, Chen X, Omari EA, Li XA et al.
cal outcomes vs photon chemo-radiotherapy. Br J Cancer. 2020;123:869–70. Development of a comprehensive cardiac atlas on a 1.5 Tesla Magnetic Reso-
80. Vinogradskiy Y, Castillo R, Castillo E, Schubert L, Jones BL, Faught A, et al. Results of a nance Linear Accelerator. Phys Imaging Radiat Oncol 2023; 28. https://doi.org/
Multi-Institutional Phase 2 Clinical Trial for 4DCT-Ventilation Functional Avoidance 10.1016/J.PHRO.2023.100504.
Thoracic Radiation Therapy. Int J Radiat Oncol*Biol*Phys. 2022;112:986–95. 102. Walls GM, Giacometti V, Apte A, Thor M, McCann C, Hanna GG, et al. Validation of
81. Hotca A, Thor M, Deasy JO, Rimner A. Dose to the cardio-pulmonary system and an established deep learning auto-segmentation tool for cardiac substructures in
treatment-induced electrocardiogram abnormalities in locally advanced non- 4D radiotherapy planning scans. Phys Imaging Radiat Oncol. 2022;23:118–26.
small cell lung cancer. Clin Transl. Radiat Oncol. 2019;19:96. 103. Kearney M, Keys M, Faivre-Finn C, Wang Z, Aznar MC, Duane F. Exposure of the
82. Miller R, Santangelo T, Forghani-Arani F, Rusthoven C, Chen Y, Castillo E, et al. heart in lung cancer radiation therapy: A systematic review of heart doses
Changes in post-treatment cardiac PET avidity predict overall survival in lung published during 2013 to 2020. Radiother Oncol. 2022;172:118–25.
cancer patients treated with chemoradiation: Secondary analysis of the ACRIN 104. McNew LK, Bowen SR, Gopan O, Nyflot MJ, Patel SA, Zeng J, et al. The relationship
6668/RTOG 0235 clinical trial. Radiother Oncol. 2022;171:22–24. between cardiac radiation dose and mediastinal lymph node involvement in stage
83. Choi W, Jia Y, Kwak J, Werner-Wasik M, Dicker AP, Simone NL et al. Novel III non-small cell lung cancer patients. Adv Radiat Oncol. 2017;2:192.
Functional Radiomics for Prediction of Cardiac Positron Emission Tomography 105. Ferris MJ, Martin KS, Switchenko JM, Kayode OA, Wolf J, Dang Q, et al. Sparing
Avidity in Lung Cancer Radiotherapy. JCO Clin Cancer Inform 2024; 8. https:// Cardiac Substructures With Optimized Volumetric Modulated Arc Therapy and
doi.org/10.1200/CCI.23.00241. Intensity Modulated Proton Therapy in Thoracic Radiation for Locally Advanced
84. Chen L, Ta S, Wu W, Wang C, Zhang Q. Prognostic and Added Value of Echo- Non-small Cell Lung Cancer. Pract. Radiat Oncol. 2019;9:e473–e481.
cardiographic Strain for Prediction of Adverse Outcomes in Patients with Locally 106. Turtle L, Bhalla N, Willett A, Biggar R, Leadbetter J, Georgiou G, et al. Cardiac-
Advanced Non-Small Cell Lung Cancer after Radiotherapy. Ultrasound Med Biol. sparing radiotherapy for locally advanced non-small cell lung cancer. Radiat
2019;45:98–107. Oncol. 2021;16:1–10.
85. Li T, Zhuang H, Wang Y, Li J, Zhu D, Cui M. Two-dimensional speckle tracking 107. Morris ED, Aldridge K, Ghanem AI, Zhu S, Glide-Hurst CK. Incorporating sensitive
echocardiography in evaluating radiation-induced heart damage. Asia Pac J cardiac substructure sparing into radiation therapy planning. J Appl Clin Med
Oncol Nurs. 2021;9:119–24. Phys. 2020;21:195–204.
86. Wang K, Malkin HE, Patchett ND, Pearlstein KA, Heiling HM, McCabe SD, et al. 108. Vasquez Osorio EM, McCallum H, Bedair A, Faivre-Finn C, Haughey A, van Herk
Coronary Artery Calcifications and Cardiac Risk After Radiation Therapy for Stage M, et al. Protecting the Heart: A Practical Approach to Account for the Full
III Lung Cancer. Int J Radiat Oncol Biol Phys. 2022;112:188–96. Extent of Heart Motion in Radiation Therapy Planning. Int J Radiat Oncol Biol
87. Atkins KM, Weiss J, Zeleznik R, Bitterman DS, Chaunzwa TL, Huynh E et al. Phys. 2020;108:1082–90.
Elevated Coronary Artery Calcium Quantified by a Validated Deep Learning 109. Aupérin A, Le Péchoux C, Pignon JP, Koning C, Jeremic B, Clamon G, et al.
Model From Lung Cancer Radiotherapy Planning Scans Predicts Mortality. JCO Concomitant radio-chemotherapy based on platin compounds in patients with
Clin Cancer Inform 2022; 6. https://doi.org/10.1200/CCI.21.00095. locally advanced non-small cell lung cancer (NSCLC): a meta-analysis of indivi-
88. Joolharzadeh P, Rodriguez M, Zaghlol R, Pedersen LN, Jimenez J, Bergom C, dual data from 1764 patients. Ann Oncol. 2006;17:473–83.
et al. Recent Advances in Serum Biomarkers for Risk Stratification and Patient 110. Eiger D, Pondé NF, Agbor-Tarh D, Moreno-Aspitia A, Piccart M, Hilbers FS, et al.
Management in Cardio-Oncology. Curr Cardiol Rep. 2023;25:133–46. Long-term cardiac outcomes of patients with HER2-positive breast cancer
89. Demissei BG, Freedman G, Feigenberg SJ, Plastaras JP, Maity A, Smith AM, et al. treated in the adjuvant lapatinib and/or trastuzumab Treatment Optimization
Early Changes in Cardiovascular Biomarkers with Contemporary Thoracic Trial. Br J Cancer. 2020;122:1453–60.
Radiation Therapy for Breast Cancer, Lung Cancer, and Lymphoma. Int J Radiat 111. Yegya-Raman N, Kegelman TP, Ho Lee S, Kallan MJ, Kim KN, Natarajan J et al.
Oncol Biol Phys. 2019;103:851–60. Death without progression as an endpoint to describe cardiac radiation effects

British Journal of Cancer

 G.M. Walls et al.
15
in locally advanced non-small cell lung cancer. Clin Transl Radiat Oncol 2023; 39. for stage III non-small cell lung cancer. Radiother Oncol. 2020;152:126–32.
https://doi.org/10.1016/J.CTRO.2023.100581. https://doi.org/10.1016/J.RADONC.2020.09.050.
112. Degens J, De Ruysscher D, Houben R, Kietselaer B, Bootsma G, Hendriks L, et al. 132. Yegya-Raman N, Ho Lee S, Friedes C, Wang X, Iocolano M, Kegelman TP, et al.
Are patients with stage III non-small cell lung cancer treated with chemor- Cardiac radiation dose is associated with inferior survival but not cardiac events
adiotherapy at risk for cardiac events? Results from a retrospective cohort study. in patients with locally advanced non-small cell lung cancer in the era of
BMJ Open. 2020;10:36492. immune checkpoint inhibitor consolidation. Radiother Oncol; 2024;190. https://
113. N Yegya-Raman, C Friedes, M Iocolano, S Feigenberg. Statin therapy, cardiac doi.org/10.1016/J.RADONC.2023.110005.
events, and survival in patients with non-small cell lung cancer receiving defi- 133. Olloni A, Brink C, Lorenzen EL, Jeppesen SS, Hoffmann L, Kristiansen C, et al.
nitive radiotherapy. Radiotherapy and Oncology 2023. Does coronary artery calcium score have an impact on overall survival for locally
114. Suero-Abreu GA, Zanni MV, Neilan TG. Atherosclerosis With Immune Checkpoint advanced non-small cell lung cancer treated with definitive radiotherapy.
Inhibitor Therapy: Evidence, Diagnosis, and Management: JACC: CardioOncol- Radiother Oncol; 2023;185. https://doi.org/10.1016/J.RADONC.2023.109719
ogy State-of-the-Art Review. JACC CardioOncol. 2022;4:598–615. 134. No HJ, Guo FB, Park NJ-I, Kastelowitz N, Rhee J-W, Clark DE, et al. Predicting Adverse
115. Walls GM, Oughton JB, Chalmers AJ, Brown S, Collinson F, Forster MD, et al. Cardiac Events After Radiotherapy for Locally Advanced Non–Small Cell Lung
CONCORDE: A phase I platform study of novel agents in combination with Cancer. JACC CardioOncol. 2023. https://doi.org/10.1016/j.jaccao.2023.08.007.
conventional radiotherapy in non-small-cell lung cancer. Clin Transl Radiat 135. Atkins KM, Zhang SC, Kehayias; Christopher, Guthier C, He J, Gasho JO, et al.
Oncol. 2020;25:61–66. Association of Cardiac Substructure Radiation Dose with Tachyarrhythmia and
116. Walls GM, O’Kane R, Ghita M, Kuburas R, McGarry CK, Cole AJ, et al. Murine Bradyarrhythmia After Lung Cancer Radiotherapy. JACC CardioOncol. 2024.
models of radiation cardiotoxicity: A systematic review and recommendations
for future studies. Radiother Oncol. 2022;173:19–31.
117. Jahng JWS, Little MP, No HJ, Loo BW, Wu JC Consequences of ionizing radiation AUTHOR CONTRIBUTIONS
exposure to the cardiovascular system. Nat Rev Cardiol 2024. https://doi.org/ Conceptualisation – GW, CR. Original draft – GW, CB, JM, SR, GH, PS, CR. Data curation
10.1038/S41569-024-01056-4. – GW. Formal analysis – GW. Reviewing & editing – GW, CB, JM, SR, GH, PS, CR
118. Boulet J, Peña J, Hulten EA, Neilan TG, Dragomir A, Freeman C et al. Statin use and
risk of vascular events among cancer patients after radiotherapy to the thorax, head,
and neck. J Am Heart Assoc 2019; 8. https://doi.org/10.1161/JAHA.117.005996.
119. Atkins KM, Bitterman DS, Chaunzwa TL, Williams CL, Rahman R, Kozono DE, et al. FUNDING INFORMATION
Statin Use, Heart Radiation Dose, and Survival in Locally Advanced Lung Cancer. The authors did not receive any direct funding for this work. GW held a Postdoctoral
Pract. Radiat Oncol. 2021;11:e459–e467. Bursary from Cancer Research UK, and a Visiting Fellow Scholarship from Fulbright
120. Walls GM, O’Connor J, Harbinson M, McCarron EP, Duane F, McCann C, et al. Ireland at the time when this was article was prepared.
Association between statin therapy dose intensity and radiation cardiotoxicity in
non-small cell lung cancer: Results from the NI-HEART study. Radiother Oncol.
2023;186:109762. COMPETING INTERESTS
121. Wang K, Pearlstein KA, Patchett ND, Deal AM, Mavroidis P, Jensen BC, et al. Heart GW – Speaker honorarium (Astra Zeneca). CB – no relevant relationships to declare.
dosimetric analysis of three types of cardiac toxicity in patients treated on dose- JM – Research support (Myocardial Solutions, Abbott Laboratories), modest
escalation trials for Stage III non-small-cell lung cancer. Radiother Oncol. consulting fees (Alnylam, Altathera, Bridgebio, Pfizer and Race Oncology). SR – no
2017;125:293–300. https://doi.org/10.1016/J.RADONC.2017.10.001. relevant relationships to declare. GH – Research support to institution (Varian Medical
122. McKenzie E, Zhang S, Zakariaee R, Guthier C V., Hakimian B, Mirhadi A, et al. Left Systems, ViewRay, Mevion Medical Systems, Siemens Helathineers), consulting fees
anterior descending coronary artery radiation dose association with all-cause (Varian), Patents/Royalties/IP (pending patent, Varian Medical Systems). PS – Speaker
mortality in nrg oncology trial RTOG 0617. Int J Radiat Oncol Biol Phys. Elsevier; honoraria (Varian Medical Systems). CR – Research support to institution (Merck,
2022;0. https://doi.org/10.1016/J.IJROBP.2022.11.033. Varian Medical Systems), consulting fees (Varian Medical Systems, AstraZeneca, EMD
123. Yu NY, Dewees TA, Voss MM, Breen WG, Chiang JS, Ding JX, et al. Cardio- Serono, Quantaras), Patents/Royalties/IP (Noninvasive imaging and treatment system
pulmonary toxicity following intensity-modulated proton therapy (IMPT) versus for cardiac arrhythmias WO 2017078757 A1; U.S. Provisional Application No. 62/598,
intensity-modulated radiation therapy (IMRT) for stage III non-small cell lung 162 Entitled System and method for determining segments for ablation), Leadership
cancer Nathan Y. Yu et al. Clin Lung Cancer. 2022;23:526–35. https://doi.org/ (Radialogica, EmpNia), Stock/Ownership (Radialogica, Quantaras, EmpNia)
10.1016/j.cllc.2022.07.017.
124. Johnson MD, Sura K, Mangona VS, Glick A, Wallace M, Ye H, et al. Matched-pair
analysis of high dose versus standard dose definitive chemoradiation for locally ADDITIONAL INFORMATION
advanced non–small-cell lung cancer. Clin Lung Cancer. Elsevier. 2017;18: Supplementary information The online version contains supplementary material
149–55. https://doi.org/10.1016/J.CLLC.2016.06.004. available at https://doi.org/10.1038/s41416-024-02888-0.
125. Han CB, Wang WL, Quint L, Xue JX, Matuszak M, Ten Haken R, et al. Pulmonary
Artery Invasion, High Dose Radiation, and Overall Survival in Patients with Non- Correspondence and requests for materials should be addressed to Gerard M. Walls.
Small Cell Lung Cancer. Int J Radiat Oncol Biol Phys. 2014;89:313 https://doi.org/
10.1016/J.IJROBP.2014.02.011. NIH Public Access. Reprints and permission information is available at http://www.nature.com/
126. Schytte T, Hansen O, Stolberg-Rohr T, Brink C. Cardiac toxicity and radiation reprints
dose to the heart in definitive treated non-small cell lung cancer. Acta Oncol.
2010;49:1058–60. https://doi.org/10.3109/0284186X.2010.504736. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims
127. Shepherd AF, Yu AF, Iocolano M, Leeman JE, Wild AT, Imber BS, et al. in published maps and institutional affiliations.
Increasing heart dose reduces overall survival in patients undergoing post-
operative radiation therapy for NSCLC. JTO Clin Res Rep. Elsevier Inc.;
2021;2:100209.
128. Lee CC, Chua GWY, Zheng H, Soon YY, Foo LL, Thiagarajan A, et al. Are heart doses
associated with survival in patients with non-small cell lung cancer who received Open Access This article is licensed under a Creative Commons
post-operative thoracic radiotherapy?: A national population-based study. Medicine Attribution 4.0 International License, which permits use, sharing,
(Baltimore). 2019;98:e17020. https://doi.org/10.1097/MD.0000000000017020. adaptation, distribution and reproduction in any medium or format, as long as you give
129. Niska JR, Hu J, Li J, Herman MG, Thorpe CS, Schild SE, et al. Using novel statistical appropriate credit to the original author(s) and the source, provide a link to the Creative
techniques to accurately determine the predictive dose range in a study of Commons licence, and indicate if changes were made. The images or other third party
overall survival after definitive radiotherapy for stage III non-small cell lung material in this article are included in the article’s Creative Commons licence, unless
cancer in association with heart dose. J Cancer Ther. 2021;12:505 https://doi.org/ indicated otherwise in a credit line to the material. If material is not included in the
10.4236/JCT.2021.129044. NIH Public Access. article’s Creative Commons licence and your intended use is not permitted by statutory
130. Cho SG, Kim YH, Park H, Park KS, Kim J, Ahn SJ, et al. Prediction of cardiac events regulation or exceeds the permitted use, you will need to obtain permission directly
following concurrent chemoradiation therapy for non-small-cell lung cancer from the copyright holder. To view a copy of this licence, visit http://
using FDG PET. Ann Nucl Med. 2022;36:439–49. https://doi.org/10.1007/S12149- creativecommons.org/licenses/by/4.0/.
022-01724-W/FIGURES/4. Springer Japan.
131. Jang BS, Cha MJ, Kim HJ, Oh S, Wu HG, Kim E, et al. Heart substructural dosi-
metric parameters and risk of cardiac events after definitive chemoradiotherapy © The Author(s) 2024

British Journal of Cancer