Radiology and Oncology  |  Ljubljana  |  Slovenia  |  www.radioloncol.com
Radiol Oncol 2025; 59(4): 510-521. doi: 10.2478/raon-2025-0043
510
research article
Cardiotoxicity in low-to-moderate 
cardiovascular risk patients undergoing anti-
HER2 therapy: a prospective cardiac magnetic 
resonance study
Sainan Cheng1, Mei Deng1, Linlin Qi1, Fenglan Li1, Jiaqi Chen1, Shulei Cui1, Yawen Wang1, 
Jianing Liu1, Yang Fan2, Lizhi Xie2, Jianwei Wang1
1  Department of Diagnostic Radiology, National Cancer Center/National Clinical Research Center for Cancer/Cancer 
Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China
2 MR Research China, GE Healthcare, Beijing Economic-Technological Development Area, Beijing, China
Radiol Oncol 2025; 59(4): 510-521.
Received 16 March 2025 
Accepted 23 May 2025
Correspondence to: Prof. Jianwei Wang, Ph.D., Department of Diagnostic Radiology, National Cancer Center/National Clinical Research 
Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, China. E-mail: dr_jianwei-
wang@163.com
Disclosure: No potential conflicts of interest were disclosed. 
This is an open access article distributed under the terms of the CC-BY license (https://creativecommons.org/licenses/by/4.0/).
Background. The study aimed to investigate cardiotoxicity among individuals undergoing anti- human epidermal 
growth factor receptor 2 (HER2) therapy with a low-to-moderate risk of cardiovascular complications. Cardiac mag-
netic resonance (CMR) imaging was employed in the investigation.
Patients and methods. HER2-positive breast cancer patients who underwent CMR examinations both before and 
during therapy (first follow-up: 3−5 months; second follow-up: 6−12 months) between January 2021 and December 
2022 were prospectively included. Each patient was evaluated for the risk of cardiovascular toxicity.
Results. Thirty-five HER2-positive breast cancer patients were included (48.86 ± 10.34 years). Eighty-nine percent of 
patients had low cardiovascular toxicity risk, and 11% had moderate cardiovascular toxicity risk. At follow-up CMR, 
nine (25.71%) patients developed cardiac dysfunction. At follow-up 1, there was a notable decrease in left ventricular 
ejection fraction, stroke volume index, cardiac output index, and absolute strain values, accompanied by higher T1 
and T2 values as well as end-systolic volume index compared to baseline (p ≤ 0.002). At follow-up 2, the T1 and T2 
values recovered to near baseline. The cardiac output index exhibited a continuous decline (p = 0.022), while other 
variables were similar (p > 0.05). Furthermore, at follow-up 1, the T1 value displayed a marked increase in patients with 
1−3 points in cardiovascular toxicity risk factors compared to those with no risk factors (p ≤ 0.043).
Conclusions. It is common for patients with low-to-moderate cardiovascular risk to experience early cardiotoxicity 
during anti-HER2 therapy. T1 mapping was a valuable approach for quantifying the specific extent of subtle tissue 
damage.
Key words: breast cancer; cardiotoxicity; cardiac magnetic resonance imaging; HER2; T1 mapping
Introduction
Advancements in breast cancer therapy, especially 
with targeted treatments, have yielded promis-
ing results for individuals diagnosed with hu-
man epidermal growth factor receptor 2 (HER2)-
positive breast cancer. While displaying remark-
able anticancer effects, the frequent incidence of 
cardiotoxicity resulting from cancer treatments 
hampers their clinical application and could com-
promise the quality of life for cancer survivors.1 
The European Society of Cardiology Guidelines 
Radiol Oncol 2025; 59(4): 510-521.
Cheng SN et al. / Cardiotoxicity in cardiac magnetic resonance study patients, undergoing anti-HER2 therapy 511
on cardio-oncology recommend a comprehensive 
evaluation of cardiovascular toxicity risk prior to 
initiating anticancer therapy, along with diligent 
cardiovascular monitoring throughout the cancer 
treatment journey.2 Early detection of changes in 
myocardial structure and function plays a piv-
otal role in preventing treatment disruptions and 
maintaining cardiac function.2-4 However, in clini-
cal practice, cardiac surveillance during anticancer 
therapies is often overlooked among breast cancer 
patients without high cardiovascular toxicity risk 
and among those who have received non-anthra-
cycline-based HER2-targeted therapy. Hence, clar-
ifying cardiac functional alterations in patients 
with low-to-moderate cardiovascular toxicity risk 
undergoing anti-HER2 therapy is crucial. This 
clarification can assist in devising more practical 
and cost-effective screening strategies.
Cardiac magnetic resonance (CMR) imaging 
plays a crucial role in monitoring cancer therapy-
related cardiotoxicity in this scenario, as it enables 
non-invasive identification of myocardial patholo-
gies such as diffuse fibrosis, oedema, and inflam-
mation using tissue characterization techniques.5 
While CMR studies have highlighted cardiac sys-
tolic function changes during anti-HER2 therapy, 
they have not assessed the cardiovascular toxicity 
risk of HER2-positive breast cancer patients.6-9 Data 
on CMR changes after anticancer therapy among 
HER2-positive patients with low-to-moderate car-
diovascular toxicity risk are limited. Therefore, 
this study aims to explore alterations in myocar-
dial structure and function among HER2-positive 
breast cancer patients presenting a low-to-moder-
ate cardiovascular toxicity risk profile employing 
CMR.
Patients and methods
Study participants
During the period spanning from January 2021 to 
December 2022, we conducted prospective enrol-
ment of patients diagnosed with HER2-positive 
breast cancer, all of whom were slated to undergo 
anti-HER2 therapy. The oncologist exercised dis-
cretion in managing anticancer therapy. The anti-
HER2 therapy regimen comprises two categories: 
single-targeted therapy (trastuzumab) and dual-
targeted therapy (trastuzumab plus pertuzumab). 
Within the dual-targeted therapy, there are further 
subdivisions: one includes anthracycline-based 
drugs (anthracycline chemotherapy followed by 
taxane plus dual-targeted therapy), and the other 
does not involve anthracycline-based drugs (tax-
ane plus carboplatin plus dual-targeted therapy). 
The treatment lasts for a cycle of 21 days. The ex-
clusion criteria comprised a history of prior anti-
cancer therapy, pre-existing symptomatic cardiac 
conditions, contraindications to undergo CMR, 
and refusal to take part in the study.
Based on recent clinical study findings, most 
instances of cancer therapy-related cardiac dys-
function (CTRCD) manifest within the initial 6 
months’ treatment. Therefore, our study was in-
cluded three CMR examinations, conducted before 
therapy (baseline), in the early stages after therapy 
initiation (at 3~5 months for follow-up 1), and in 
the mid-to-late stages (at 6~12 months for follow-
up 2). Those who underwent at least one follow-
up CMR examination were eligible for inclusion in 
the study. The study obtained approval from the 
hospital’s research ethics committee (Approval 
No. NCC2020C-508) and all participants furnished 
written consent with full understanding.
According to the HFA-ICOS (Heart Failure 
Association–International Cardio-Oncology 
Society) cardiovascular toxicity risk stratification, 
a low-risk level is characterized by the absence of 
risk factors or the presence of one moderate risk 
factor, each of which is assigned 1 point, while a 
moderate-risk level is defined as having moder-
ate risk factors totalling 2~4 points.2 The baseline 
1-point cardiovascular toxicity risk factor includ-
ing elevated baseline cardiac biomarkers with an-
thracycline chemotherapy, hypertension, chronic 
kidney disease, diabetes mellitus, anthracycline 
before HER2-targeted therapy, previous exposure 
to non-anthracycline chemotherapy, current smok-
er or significant smoking history, obesity (BMI > 30 
kg/m2). The baseline 2-point cardiovascular toxici-
ty risk factors including previous arrhythmia with 
HER2-targeted therapies, left ventricular ejection 
fraction (LVEF) 50−54%, elevated baseline cardiac 
biomarkers with HER2-targeted therapies, age 
65−79 years, previous exposure to anthracycline, 
previous exposure to radiotherapy to left chest or 
mediastinum.
Symptomatic CTRCD indicates heart failure, 
a clinical syndrome characterized by primary 
symptoms like breathlessness, ankle swelling, and 
fatigue. Asymptomatic CTRCD is categorized into 
three levels: (1) mild, indicated by a LVEF of 50% or 
higher, alongside a notable decrease in global lon-
gitudinal strain (GLS) exceeding 15%; (2) moder-
ate, indicated by a new decrease in LVEF of at least 
10 percentage points resulting in an LVEF ranging 
from 40–49%, or a decrease in LVEF of less than 
Radiol Oncol 2025; 59(4): 510-521.
Cheng SN et al. / Cardiotoxicity in cardiac magnetic resonance study patients, undergoing anti-HER2 therapy512
10 percentage points resulting in an LVEF ranging 
from 40–49%, accompanied by a new decline in 
GLS exceeding 15%; (3) severe, characterized by a 
new reduction in LVEF to below 40%.2,10
CMR protocols and image analysis
Scanning was conducted utilizing a 3.0 T scanner 
(SIGNA Architect, GE Healthcare, Waukesha, WI, 
USA). We acquired images by employing breath-
holding techniques. Cine images were acquired 
in LV 2-chamber, 4-chamber, LV outflow tract and 
multisection stacks of short-axis views to perform 
LV function and volumes. Oedema evaluation 
was conducted using T2-weighted dual inversion-
recovery imaging in the mid-LV short-axis orien-
tation. For T1 mapping, a modified look-locker 
inversion-recovery sequence was employed, cap-
turing images in a 3(3)3(3)5 heartbeat pattern. To 
conduct T2 mapping, a multi-echo fast spine echo 
sequence utilizing double inversion-recovery was 
utilized, featuring four distinct echo times: 10.8 
ms, 32.4 ms, 54.0 ms, and 75.5 ms. Both T1 and T2 
mapping sequences were acquired in the mid-LV 
short-axis orientation. 
CMR parameters were calculated with the CVI42 
software, developed by Circle Cardiovascular 
Imaging Inc. in Calgary, Canada, by a radiologist 
with eight years of expertise in CMR. The radi-
ologist was blinded to patients’ clinical data dur-
ing the analysis process. LV volumes parameters 
including LVEF, end-diastolic and end-systolic 
volumes, cardiac output, stroke volume and my-
ocardial mass. Variables were adjusted for body 
surface area and expressed as indexes. Parameters 
(except for ejection fraction) were calibrated based 
on body surface area and presented as indices. 
We measured the T1 and T2 values at the mid-LV 
septum in the short-axis plane. GLS was mainly 
derived from LV 2-chamber and 4-chamber view. 
Global radial strain (GRS) and global circumferen-
tial strain (GCS) were calculated using short-axis 
imaging. For intraobserver reproducibility test, the 
same radiologist re-examined measurements in 
12 randomly sampled patients after a one-month 
interval. For interobserver reproducibility test, a 
second observer, unaware of the patients’ clinical 
data, independently analysed scans from 12 ran-
domly chosen patients.
Statistical analysis
Continuous variables are displayed as means ± 
standard deviation. Categorical variables are dis-
played as frequencies and percentages. The anal-
ysis encompassed comparing measurements at 
two different time points, utilizing a paired test 
for continuous variables demonstrating normal 
distribution and employing the Wilcoxon signed-
rank test for variables without a normal distribu-
tion. Comparison between groups (patients with 
and without risk factors and patients treated with 
anthracycline-based and non-anthracycline-based 
anti-HER2 therapies) were assessed using either 
the unpaired t-test or the Mann‒Whitney U test. 
Statistical significance was established for p-val-
ues less than 0.05. Statistical analyses were con-
ducted using SPSS Statistics version 26 (IBM) and 
Prism version 9.3.0 (GraphPad Software).
Results 
Patient characteristics and cancer 
therapy-related cardiac dysfunction 
(CTRCD)
35 female patients diagnosed with HER2-positive 
breast cancer were ultimately included (mean age 
± standard deviation, 48.86 years ± 10.34). Of the 
35 patients, 17 (48%) received taxane plus carbopl-
atin plus trastuzumab plus pertuzumab (TCbHP), 
8 (23%) received taxane plus carboplatin plus tras-
tuzumab (TCbH) and 10 (29%) received anthra-
cycline chemotherapy followed by a taxane plus 
trastuzumab plus pertuzumab (AC/EC-THP) treat-
ment (Figure 1). Five (14%) patients had hyperten-
sion, four were obese and two were above the age 
of 65 at baseline. All patients underwent an elec-
trocardiogram examination before treatment, but 
for young patients with no prior medical history, 
cardiac biomarker tests were not performed. There 
were no apparent abnormalities in the patient’s 
baseline electrocardiogram. All patients were non-
smokers with no significant smoking history, and 
none had chronic kidney disease. 89% of the pa-
tients exhibited a low-risk level for cardiovascu-
lar toxicity, while 11% presented a moderate-risk 
level. Table 1 presents the foundational traits at the 
study’s outset.
Thirty-two patients underwent cardiac MR 
scans at follow-up 1, with a mean time interval of 
102.22 ± 20.58 days. Sixteen patients underwent 
cardiac MR scans at follow-up 2, with a mean time 
interval of 242.81 ± 68.29 days. Thirteen patients 
completed cardiac MR scans at both follow-up 1 
and follow-up 2. For patients receiving both tar-
geted therapy and anthracycline drugs, targeted 
therapy was administered sequentially after the 
Radiol Oncol 2025; 59(4): 510-521.
Cheng SN et al. / Cardiotoxicity in cardiac magnetic resonance study patients, undergoing anti-HER2 therapy 513
completion of anthracycline treatment, with no in-
stances of simultaneous usage. 
We observed a significant increase in heart rate 
at follow-up 1 compared to baseline. However, no 
significant changes were detected in other vital 
signs, such as blood pressure or respiratory rate, 
during the same period.
A total of nine (25.71%) patients had experi-
enced CTRCD during the follow-up CMR scans. 
Five of these patients showed no cardiovascular 
toxicity risk factors, while four of them showed 1−3 
points risk factors. 
One patient developed symptomatic CTRCD 
at follow-up 1, classified as New York Heart 
Association functional class III, with fatigue and 
dyspnoea as the primary symptoms. This was ac-
companied by a decrease in LVEF from 56% to 35%, 
a reduction in GLS from -11.18 to -9.48, a decrease 
in GRS from 18.65 to 15.78, a decrease in GCS from 
-12.76 to -11.42, an increased in native T1 value 
from 1239 msec to 1271 msec, and an increased in 
T2 value from 47.90 msec to 48.57 msec. The pa-
tient sought treatment at a hospital outside the 
local area, where she was treated with the tradi-
tional Chinese medicine for heart failure and also 
received antihypertensive therapy. Subsequently, 
the patient discontinued breast cancer treatment 
and did not return to our hospital for further fol-
low-up.
Seven out of the nine patients exhibited mild 
asymptomatic CTRCD (six at follow-up 1 and one 
at follow-up 2), while one patient demonstrated 
moderate asymptomatic CTRCD at follow-up 2. 
Among the nine patients with CTRCD, three re-
ceived anthracycline-based chemotherapy fol-
lowed by taxane and dual-targeted therapy. Four 
received taxane in combination with carboplatin 
and dual-targeted therapy, while two were treated 
with taxane, carboplatin and trastuzumab. One 
patient with mild asymptomatic CTRCD had a 
history of hypertension for over 20 years and had 
been on long-term telmisartan therapy. None of the 
other patients received specific guideline-directed 
medical therapy for heart failure, such as ACE in-
hibitors, beta-blockers, or mineralocorticoid recep-
tor antagonists, as recommended in the 2022 ESC 
cardio-oncology guidelines. Two representative 
clinical cases are shown in Figures 2 and 3. 
FIGURE 1. Study flow diagram.
HER2 = human epidermal growth factor receptor 2
Radiol Oncol 2025; 59(4): 510-521.
Cheng SN et al. / Cardiotoxicity in cardiac magnetic resonance study patients, undergoing anti-HER2 therapy514
tributed to the increase in LV end-systolic volume 
(24.53 mL/m² ± 6.17 vs. 28.26 mL/m² ± 9.15; p = 0.001), 
with no significant changes observed in end-dias-
tolic volume (p = 0.057). The LV mass index showed 
no significant statistical difference (p = 0.369).
We observed significantly decreases in absolute 
myocardial strain values: 33.59% ± 7.31 vs. 28.81% 
± 6.37 for GRS, -18.86% ± 2.41 vs. -17.38% ± 2.31 for 
GCS, and -16.12% ± 1.55 vs. -15.09% ± 1.85 for GLS; 
all with a p-value of ≤ 0.001).
T2-weighted short-tau inversion-recovery imag-
ing did not reveal apparent myocardial oedema. 
However, we noted significant increases in both 
T1 (1231.16 msec ± 46.49 vs. 1255.81 msec ± 45.23; 
p = 0.001) and T2 values (47.86 msec ± 2.17 vs. 49.43 
msec ± 2.71; p = 0.001) (Table 2).
Differences in cardiac magnetic 
resonance (CMR) results before 
treatment and during the mid-to-late 
stage after treatment
Similarly to the aforementioned findings, there 
were significant decreases observed in LVEF, 
stroke volume index, cardiac output index, and 
absolute strain parameter values from baseline 
to follow-up 2 (p < 0.05). Conversely, the LV end-
systolic volume index showed a notable increase (p 
= 0.002). Noteworthy, no substantial changes were 
detected in T1 and T2 values (p > 0.05) (Table 2, 
Figure 4).
Differences in cardiac magnetic 
resonance (CMR) results between two 
subsequent follow-up scans
The average time span between the two subse-
quent follow-up CMR scans was 136.92 ± 67.70 
days. A notable decrease in cardiac output index 
was observed (3.01 L/min×m2 ± 0.47 vs. 2.60 L/
min×m2 ± 0.54; p = 0.022). No statistically notable 
differences were detected in the remaining vari-
ables (p > 0.05) (Table 3).
Differences in cardiac magnetic 
resonance (CMR) results between 
patients with and without risk factors at 
the early stage after treatment
Thirty-two patients completed cardiac MR scans 
at follow-up 1, 16 of whom had no cardiovascu-
lar toxicity risk factors and 16 of whom had 1-3 
risk factors. Patients presenting with 1-3 risk fac-
tors exhibited significantly higher LV mass index 
TABLE 1. Clinical characteristics and treatment of the study 
participants
Variable Value
  No. of participants 35
  Age, years* 48.86 ± 10.34
  Height (cm)* 160.97 ± 4.20
  Weight (kg)* 62.61 ± 8.36
  Body surface area (m2)* 1.63 ± 0.12
  Hypertension† 5 (14%)
  Chronic kidney disease 0
   Current smoker or significant smoking 
history 0
  Obesity (BMI > 30 kg/m2) † 4 (11%)
  Age 65−79 years† 2 (6%)
  Diabetes 0
  Hyperlipidaemia 0
  Prior cardiovascular disease 0
   HFA-ICOS baseline cardiovascular 
toxicity risk stratification†
   Moderate risk (moderate risk factors 
with a total of 2−3 points) 4 (11%)
  Low risk 31 (89%)
  No risk factor 18 (52%)
   One moderate risk factor with a total 
of 1 point 13 (37%)
  Therapies†
   HER2-targeted therapies without 
anthracycline 25 (71%)
  Trastuzumab 8 (23%)
   Trastuzumab and pertuzumab 17 (48%)
   Anthracycline chemotherapy 
followed by a taxane plus
  trastuzumab plus pertuzumab
10 (29%)
* Data are given as means ± standard deviations. †Data in parentheses 
are percentages.
BMI = body mass index; HFA-ICOS = Heart Failure Association-
International Cardio-Oncology Society; HER2 = human epidermal 
growth factor receptor 2
Differences in cardiac magnetic 
resonance (CMR) results before 
treatment and during the early stage 
after treatment
At follow-up 1, the LV systolic function showed 
a significant decrease compared to the baseline 
measurement, including reductions in LVEF 
(64.03% ± 4.16 vs. 57.34% ± 5.32; p < 0.001), stroke 
volume index (43.36 mL/m² ± 8.12 vs. 37.10 mL/m² 
± 6.43; p < 0.001) and cardiac output index (3.23 L/
min×m2 ± 0.58 vs. 2.91 L/min×m2 ± 0.49; p = 0.002). 
The decline in LV systolic function is primarily at-
Radiol Oncol 2025; 59(4): 510-521.
Cheng SN et al. / Cardiotoxicity in cardiac magnetic resonance study patients, undergoing anti-HER2 therapy 515
(37.07 g/m2 ± 6.68 vs. 43.03 g/m2 ± 8.86, p = 0.040) 
and T1 value (1235.06 ± 36.77 vs. 1276.56 msec ± 
44.25, p = 0.007) (Figure 5). However, no statisti-
cally significant disparity in LV mass index or T1 
value was observed between these two groups 
at baseline. The two groups also had similar LV 
ejection fraction, volume index, cardiac output 
index, strain parameters and T2 values (p > 0.05) 
(Table 4).
Differences in cardiac magnetic 
resonance (CMR) results between 
patients with and without anthracycline 
therapy at the early stage after 
treatment
Ten patients were taking anthracycline-based 
therapy and 22 patients were taking non-anthra-
cycline-based therapy at follow-up 1. The group 
treated with anthracycline showed a notably el-
evated T1 value (1280.90 msec ± 43.10 vs. 1244.41 
msec ± 42.29, p = 0.032). At baseline, the T1 value 
did not show any significant variance between the 
two groups. The two groups were similar in all 
other variables (p > 0.05) (Table 5).
Discussion
In this research, our aim was to explore cardiotox-
icity among individuals undergoing anti-HER2 
therapy with a low-to-moderate risk of cardio-
vascular complications. The main findings are as 
follows: (1) Subclinical CTRCD was frequently ob-
served in low-to-moderate cardiovascular toxicity 
risk patients with anti-HER2 therapy. (2) CTRCD 
was detected as early as the third month of anti-
cancer therapy and persisted for at least approxi-
mately eight months. (3) After anticancer therapy, 
patients with 1−3 cardiovascular toxicity risk 
points, including those who underwent anthra-
cycline-based therapy, exhibited higher T1 values 
compared to those without such risk factors, while 
showing no notable differences in LVEF or myo-
cardial strain values.
Breast cancer patients categorized with low-to-
moderate cardiovascular toxicity risk demonstrate 
an extended life expectancy relative to those with 
high cardiovascular toxicity risk, potentially allow-
ing for the consideration of more aggressive target-
ed therapy and chemotherapy interventions.10 In 
our study, 89% of breast cancer patients were at low 
TABLE 2. Cardiac magnetic resonance (CMR) parameter results at baseline and follow-up
Variable
Follow-up 1 (n = 32) Follow-up 2 (n = 16)
Baseline 1 Follow-up 1 P Value Baseline 2 Follow-up 2 P Value
LV ejection fraction 
(%) 64.03 ± 4.16 57.34 ± 5.32 < 0.001 66.23 ± 4.19 57.45 ± 5.54 < 0.001
LV end-diastolic 
volume index  
(mL/m2)
68.23 ± 12.41 65.63 ± 12.97 > 0.057 68.20 ± 11.56 64.94 ± 14.01 0.275
LV end-systolic 
volume index  
(mL/m2)
24.53 ± 6.17 28.26 ± 9.15 0.001 22.91 ± 5.02 27.70 ± 7.60 0.002
LV mass index (g/m2) 37.17  (33.96, 43.69)
38.24  
(34.40, 41.45) 0.369 39.23 ± 6.31 40.03 ± 6.78 0.261
Stroke volume index 
(mL/m2) 43.36 ± 8.12 37.10 ± 6.43 < 0.001 44.84 ± 8.12
39.46  
(30.13, 42.17) 0.002
Cardiac output index 
(L/min×m2) 3.23 ± 0.58 2.91 ± 0.49 0.002 3.19 ± 0.38 2.65 ± 0.51 0.001
Global radial strain 33.59 ± 7.31 28.81 ± 6.37 0.001 37.16 ± 8.99 30.07 ± 6.18 < 0.001
Global 
circumferential strain -18.86 ± 2.41 -17.38 ± 2.31 < 0.001 -19.93 ± 2.81 -17.23 ± 2.34 < 0.001
Global longitudinal 
strain -16.12 ± 1.55 -15.09 ±1.85 0.001 -16.67 ± 1.28 -14.88 ± 1.60 0.004
T1 value (msec) 1231.16  46.49 1255.81 ± 45.23 0.001 1222.00 ± 40.89 1237.13 ± 35.19 0.051
T2 value (msec) 47.86 ± 2.17 49.43 ± 2.71 0.001 48.13 ± 2.15 48.84 ± 1.75 0.306
Heart rate during MR 
scan (bpm) 74.50 ± 11.11 79.47 ± 13.40 0.048 72.50 ± 9.68 72.69 ± 12.28 0.928
Data are given as means ± standard deviations or medians with interquartile ranges in parentheses. Baseline 1 and 2 were calculated separately for patients who 
underwent follow-up exams 1 and 2.
bpm = beats per minute; LV = left ventricular
Radiol Oncol 2025; 59(4): 510-521.
Cheng SN et al. / Cardiotoxicity in cardiac magnetic resonance study patients, undergoing anti-HER2 therapy516
FIGURE 2. CMR imaging in a 44-year-old woman with human epidermal growth factor receptor 2 (HER2)-positive right-side 
breast cancer. The top row images (A, B, C, D) represent the baseline. The middle row images (E, F, G, H) correspond to the 
condition after 4 cycles of the TCbHP regimen (104 days after the baseline scan). The bottom row images (I, J, K, L) show the 
status after 9 cycles of the TCbHP regimen (203 days after the baseline scan). In the axial breast enhancement scan images 
(A, E, I), a decrease in the tumour size of the right breast is observed (arrow). The diastolic phase images of left ventricular 
two-chamber plane (B, F, J) demonstrate a slight increase in left ventricular systolic volume. T1 mapping images (C, G, I) reveal 
an increase in T1 value in the septum of the mid short-axis slice during follow-up 1, followed by a decrease during follow-up 
2. The global radial, circumferential and longitudinal strain (GRS, GCS and GLS) curves before and after TCbHP treatment are 
presented (D, H, L).
CMR = cardiac magnetic resonance; GCS = global circumferential strain; GLS = global longitudinal strain; GRS = global radial strain; HER2 = human 
epidermal growth factor receptor 2; TCbHP = taxane plus carboplatin plus trastuzumab plus pertuzumab
FIGURE 3. CMR imaging in a 68-year-old woman with HER2-positive right-side breast cancer. The upper images (A, B, C, D) 
depict the baseline condition. The lower images (E, F, G, H) represent the status after 4 cycles of the TCbHP regimen (106 days 
following the baseline scan). Axial breast enhancement scan images (A, E) exhibit a reduction in the tumour size of the right 
breast. Short-axis cine images reveal an enlargement in the left ventricular diastolic diameter (B, F). T1 mapping images (C, 
G) reveal an increase in T1 value in the septum of the mid short-axis slice during the follow-up. Strain curves demonstrate a 
reduction in the absolute values of global radial, circumferential and longitudinal strain (GRS, GCS and GLS) before and after 
treatment (D, H).
CMR = cardiac magnetic resonance; GCS = global circumferential strain; GLS = global longitudinal strain; GRS = global radial strain; HER2 = human 
epidermal growth factor receptor 2; TCbHP = taxane plus carboplatin plus trastuzumab plus pertuzumab
A
A
B
B
C
C
D
D
E
E
F
F
J K L
G
G
H
H
I
Radiol Oncol 2025; 59(4): 510-521.
Cheng SN et al. / Cardiotoxicity in cardiac magnetic resonance study patients, undergoing anti-HER2 therapy 517
TABLE 3. Cardiac magnetic resonance (CMR) parameter results between follow-up 1 and follow-up 2 (n = 13)
Variable Follow-up 1 Follow-up 2 P Value
LV ejection fraction (%) 58.45 ± 4.44 57.76 ± 5.53 0.545
LV end-diastolic volume index (mL/m2) 66.33 ± 11.13 62.32 ± 14.02 0.055
LV end-systolic volume index (mL/m2) 27.59 ± 6.20 26.35 ± 7.46 0.294
LV mass index (g/m2) 38.07 (36.22, 44.09) 39.13 (35.77, 43.31) 0.650
Stroke volume index (mL/m2) 39.37 (35.27, 43.55) 39.44 (27.44, 40.83) 0.101
Cardiac output index (L/min×m2) 3.01 ± 0.47 2.60 ± 0.54 0.022
Global radial strain 28.69 ± 8.12 29.70 ± 6.48 0.670
Global circumferential strain -17.70 ± 2.52 -17.12 ± 2.30 0.172
Global longitudinal strain -15.01 ± 1.46 -14.74 ±1.70 0.647
T1 value (msec) 1245.31 ± 22.46 1236.77 ± 40.36 0.486
T2 value (msec) 49.67 ± 2.19 49.11 ± 2.12 0.465
Heart rate during MR scan (bpm) 78.38 ± 14.67 73.92 ± 13.09 0.189
Data are given as means ± standard deviations or medians with interquartile ranges in parentheses. 
bpm = beats per minute; CMR = cardiac magnetic resonance; LV = left ventricular
FIGURE 4. Box plot show changes in cardiac magnetic resonance (CMR) variables between the baseline and follow-up CMR 
examinations. The results of CMR variables at baseline (n = 35), follow-up 1 (n = 32) and follow-up 2 (n = 16) were shown in blue, 
yellow and red box plot. Left ventricular ejection fraction (LVEF) decreased significantly at follow-up 1 and follow-up 2. T1 and 
T2 value increased significantly at follow-up 1, but recovered to a level not significantly different from the baseline. Global 
longitudinal strain (GLS), global radial strain (GRS) and global circumferential strain (GCS) decreased significantly at follow-up 
1 and follow-up 2.
Radiol Oncol 2025; 59(4): 510-521.
Cheng SN et al. / Cardiotoxicity in cardiac magnetic resonance study patients, undergoing anti-HER2 therapy518
cardiovascular toxicity risk, and 52% had no risk 
factors. As demonstrated in Table 4, the presence 
of risk factors did not appear to significantly influ-
ence CMR parameters at follow-up 1. Patients with 
1−3 risk factors exhibited higher LV mass index 
and native T1 values, whereas other parameters, 
such as LV ejection fraction, volume index, cardiac 
output index, strain parameters and T2 values, re-
mained comparable between groups. These find-
ings suggest that monitoring cardiac function dur-
ing anticancer therapy is equally important even in 
patients with low cardiovascular toxicity risk.
In our study, 25.71% of patients developed 
CTRCD during treatment, as monitored through 
TABLE 4. Differences in CMR parameters at follow-up 1 between patients with and without baseline risk factors
Variable Patients with no risk factor(n = 16)
Patients with 1-3 risk 
factors (n = 16) P Value
LV ejection fraction (%) 57.92 ± 3.43 56.75 ± 6.79 0.543
LV end-diastolic volume index (mL/m2) 63.88 ± 10.89 67.39 ± 14.92 0.454
LV end-systolic volume index (mL/m2) 26.89 ± 5.66 29.63 ± 11.70 0.406
LV mass index (g/m2) 37.07 ± 6.68 43.03 ± 8.86 0.040
Stroke volume index (mL/m2) 36.60 ± 6.67 37.61 ± 6.36 0.665
Cardiac output index (L/min×m2) 2.84 ± 0.48 2.99 ± 0.49 0.387
Global radial strain 30.56 ± 5.40 27.07 ± 6.94 0.123
Global circumferential strain -17.94 ± 2.16 -16.82 ± 2.37 0.174
Global longitudinal strain -15.11 ± 1.85 -15.06 ± 1.90 0.938
T1 value (msec) 1235.06 ± 36.77 1276.56 ± 44.25 0.007
T2 value (msec) 48.95 ± 2.28 49.94 ± 3.07 0.318
Heart rate during MR scan (bpm) 77.56 ± 11.28 81.38 ± 15.37 0.430
Data are given as means ± standard deviations or medians with interquartile ranges in parentheses.
CMR = cardiac magnetic resonance; LV = left ventricular
TABLE 5. CMR parameter results between patients treated with anthracycline-based and non-anthracycline-based anti- 
human epidermal growth factor receptor 2 (HER2) therapies at follow-up 1
Variable Anthracycline-based therapy (n = 10)
Non-anthracycline-based 
therapy (n = 22) P Value
LV ejection fraction (%) 57.01 ± 3.68 57.49 ± 5.99 0.816
LV end-diastolic volume index (mL/m2) 67.51 ± 12.83 64.78 ± 13.24 0.590
LV end-systolic volume index (mL/m2) 28.88 ± 6.01 27.98 ± 10.38 0.800
LV mass index (g/m2) 41.76 ± 7.91 39.28 ± 8.53 0.442
Stroke volume index (mL/m2) 38.58 ± 7.78 36.43 ± 5.80 0.389
Cardiac output index (L/min×m2) 3.13 ± 0.49 2.82 ± 0.46 0.094
Global radial strain 26.73 ± 6.62 29.76 ± 6.16 0.218
Global circumferential strain -17.09 ± 1.92 -17.52 ± 2.49 0.637
Global longitudinal strain -15.36 (-16.56, -13.74) -15.07 ± 2.05 1.000
T1 value (msec) 1280.90 ± 43.10 1244.41 ± 42.29 0.032
T2 value (msec) 49.65 (47.06, 53.10) 49.04 ± 2.23 0.231
Heart rate during MR scan (bpm) 84.20 ± 18.44 77.32 ± 10.20 0.182
Data are given as means ± standard deviations or medians with interquartile ranges in parentheses. 
CMR = cardiac magnetic resonance; LV = left ventricular
Radiol Oncol 2025; 59(4): 510-521.
Cheng SN et al. / Cardiotoxicity in cardiac magnetic resonance study patients, undergoing anti-HER2 therapy 519
CMR examinations. Among the CTRCD patients, 
more than half had no risk factors before anti-
cancer treatment. Despite the relatively low-to-
moderate risk profile of our study population, 
the determined rate of CTRCD aligned with the 
ranges in earlier studies11,12 and was similar to a re-
cent result.13 Thus, this is a critical group to moni-
tor for cardiac toxicity during anticancer therapy. 
Regular evaluation of LV function, utilizing meas-
urements of LVEF and GLS, is advised before start-
ing HER2-targeted therapy and should continue 
every three months during the treatment moni-
toring period.2,14 In the case of asymptomatic mild 
CTRCD within this patient cohort, the guidelines 
propose the consideration of cardioprotective ther-
apy (angiotensin-converting enzyme inhibitors, 
angiotensin receptor blockers and beta-adrenergic 
blockers). Additionally, it’s recommended to con-
tinue HER2-targeted treatment.2,14-16 However, car-
diac surveillance is not always strictly performed 
in real-world clinical practice. In a clinical study 
encompassing 4,325 patients, it was observed that 
just 46.2% of breast cancer patients receiving tras-
tuzumab underwent the suggested cardiac sur-
veillance.17 Among low cardiovascular toxicity 
risk patients, the frequency of cardiac surveillance 
might be even less frequent in actual practice. For 
low cardiovascular toxicity risk patients, monitor-
ing cardiac function during anticancer therapy 
remains equally important, but perhaps the moni-
toring frequency could be appropriately reduced.
In our study, cardiac dysfunction was observed 
at the third month of anticancer therapy, and this 
significant decline persisted into the eighth month. 
Cardiac motion function did not change signifi-
cantly between the third and eighth months. 
Notably, there was a substantial rise in T1 and 
T2 values noted at the early stage after treatment, 
but by the mid-to-late stage after treatment, these 
values had reverted to a level that did not show 
significant variance compared to the baseline. In 
essence, cardiotoxicity manifested as early as the 
third month of anticancer therapy, but for the ma-
jority of patients, its progression did not continue 
beyond approximately the eighth month. Our 
study aligns with findings from prior research. 
Houbois et al.’s findings suggest a decrease in 
both LVEF and strain values over the follow-up 
duration, with the lowest point reached at three 
months into trastuzumab therapy.7 At both 6 and 
12 months, Gong et al. observed noteworthy de-
clines in systolic function; however, these dispari-
ties ceased to be significant by the 18-month mark.9 
Recent analyses on real-world data demonstrated 
that the time of onset of HER2 inhibitor-related 
cardiotoxicity was 80.50~103.00 days.18,19 This find-
ing holds significance, indicating that among in-
dividuals diagnosed with HER2-positive breast 
cancer and presenting mild-to-moderate cardio-
vascular toxicity risk factors, conducting early and 
less frequent cardiotoxicity screenings could po-
tentially improve cost-effectiveness.
Another significant observation from this study 
was that HER2-positive breast cancer patients with 
1−3 cardiovascular toxicity risk factors displayed a 
minor yet statistically significant elevation in na-
tive T1 value following anticancer therapy, in com-
parison to those without any risk factors. However, 
there were no notable variations found in terms of 
statistical significance concerning LV ejection frac-
tion or myocardial strain values. Typically, a rise 
in native T1 values correlates with myocardial 
conditions such as oedema, inflammation, and fi-
brosis20, which is able to reflect subtle myocardial 
histological changes. A previous clinical study has 
demonstrated that early cardiac changes, there 
was a progressive increase in native T1 and T2 
values. Subsequently, as treatment progressed, 
there was a normalization of T2 time, while non-
contrast T1 values remained persistently elevated, 
indicating the presence of myocardial fibrosis.21 
The detection rate of cardiac abnormalities after 
cancer therapy was 87% based on T1 and T2 map-
FIGURE 5. The comparison of T1 value between the 
baseline and follow-up cardiac magnetic resonance (CMR) 
examinations. Bars represent median T1 value. Blue dots 
represent human epidermal growth factor receptor 2 (HER2)-
positive breast cancer patients with no cardiovascular 
toxicity risk factor and orange dots represent patients with 
1−3 risk factors.
Radiol Oncol 2025; 59(4): 510-521.
Cheng SN et al. / Cardiotoxicity in cardiac magnetic resonance study patients, undergoing anti-HER2 therapy520
ping. In comparison, abnormalities based on GLS 
and LVEF were found in 71% and 66% of patients 
respectively.21 Therefore, in identifying myocardi-
al alterations among HER2-positive breast cancer 
patients following anticancer therapy, T1 mapping 
is a favoured method for precisely measuring the 
particular degree of tissue damage. This specific 
damage might not be easily identified through tra-
ditional cine sequences or strain analysis.
During the follow-up period, we noticed a no-
table increase in native T1 values among patients 
who underwent anthracycline-based therapy com-
pared to those who underwent non-anthracycline-
based therapy. This difference may be attributed 
to the potential of anthracycline chemotherapy to 
induce acute, subacute, and chronic myocellular 
injuries. Our findings align with a previous study, 
demonstrating increased T1 value in individuals 
with prior anthracycline chemotherapy exposure.22
There were several limitations acknowledged in 
this study. Firstly, the sample size was compara-
tively limited. This prospective study aimed to 
compare CMR parameters before and after anti-
cancer treatment, illustrating the temporal chang-
es in these parameters over time. Consequently, 
only patients with at least two CMR scans were 
included, underscoring the necessity for a study 
with a larger sample size. Secondly, the follow-up 
duration for CMR monitoring was short, which 
hindered the assessment of the enduring cardiac 
repercussions of anti-HER2 therapy. This area 
requires further investigation in future research. 
Thirdly, cardiac biomarkers included in patients’ 
haematological tests were not used for cardiovas-
cular toxicity risk stratification prior to treatment, 
which may have introduced some bias. However, 
most patients in our cohort did not receive anthra-
cycline-based chemotherapy and were relatively 
young, without a history or symptoms of cardio-
vascular disease, making abnormal biomarker lev-
els unlikely. Therefore, the impact of this limitation 
was expected to be minimal, although it would be 
addressed more thoroughly in future studies.
In conclusion, asymptomatic cardiac dysfunc-
tion was common among HER2-positive breast 
cancer patients with low-moderate cardiovascular 
toxicity risk after cancer therapy. This subgroup 
of patients can potentially benefit from early post-
anticancer therapy monitoring at longer intervals, 
thus enhancing cost-effectiveness. T1 mapping is a 
valuable approach for quantifying the specific ex-
tent of tissue damage, particularly in the presence 
of subtle tissue changes. There’s a need for stud-
ies that encompass larger cohorts and extended 
follow-up durations.
Acknowledgments
This study has received funding by National 
Natural Science Foundation of China (Grant 
No. 82402230, 81971616), Beijing Natural Science 
Foundation (Grant No.7222148), Medical and 
Health Science and Technology Innovation Project 
of Chinese Academy of Medical Sciences (Grant 
No. 2021-I2M-C&T-B-065), Beijing Hope Run 
Special Fund of Cancer Foundation of China (Grant 
No.LC2020B19) and National High Level Hospital 
Clinical Research Funding and Cooperation Fund 
of CHCAMS and SZCH Fund (CFA202202016).
References
1. Herrmann J. Adverse cardiac effects of cancer therapies: cardiotoxicity and 
arrhythmia. Nat Rev Cardiol 2020; 17: 474-502. doi: 10.1038/s41569-020-
0348-1
2. Lyon AR, Lopez-Fernandez T, Couch LS, Asteggiano R, Aznar MC, Bergler-
Klein J, et al. 2022 ESC Guidelines on cardio-oncology developed in 
collaboration with the European Hematology Association (EHA), the 
European Society for Therapeutic Radiology and Oncology (ESTRO) and 
the International Cardio-Oncology Society (IC-OS). Eur Heart J Cardiovasc 
Imaging 2022; 23: e333-e465. doi: 10.1093/ehjci/jeac106
3. Thavendiranathan P, Negishi T, Somerset E, Negishi K, Penicka M, Lemieux J, 
et al. Strain-guided management of potentially cardiotoxic cancer therapy. J 
Am Col Cardiol 2021; 77: 392-401. doi: 10.1016/j.jacc.2020.11.020
4. Goulas K, Farmakis D, Constantinidou A, Kadoglou NPE. Cardioprotective 
agents for the primary prevention of trastuzumab-associated cardiotoxicity: 
a systematic review and meta-analysis. Pharmaceuticals 2023; 16: 983. doi: 
10.3390/ph16070983
5. Taylor AJ, Salerno M, Dharmakumar R, Jerosch-Herold M. T1 mapping: basic 
techniques and clinical applications. JACC: Cardiovas Imag 2016; 9: 67-81. 
doi: 10.1016/j.jcmg.2015.11.005 
6. Cheng S, Wang J, Wang Y, Qi L, Li F, Liu J, et al. Longitudinal assessment 
of cardiac parameters through MRI in breast cancer patients treated with 
anti-HER2 therapy. Eur Radiol Exper 2023; 7: 1-9. doi: 10.1186/s41747-023-
00338-9
7. Houbois CP, Nolan M, Somerset E, Shalmon T, Esmaeilzadeh M, Lamacie 
MM, et al. Serial cardiovascular magnetic resonance strain measure-
ments to identify cardiotoxicity in breast cancer: comparison with echo-
cardiography. JACC Cardiovasc Imaging 2021; 14: 962-74. doi: 10.1016/j.
jcmg.2020.09.039
8. Gong IY, Ong G, Brezden-Masley C, Dhir V, Deva DP, Chan KKW, et al. Early 
diastolic strain rate measurements by cardiac MRI in breast cancer patients 
treated with trastuzumab: a longitudinal study. Int J Cardiovasc Imaging 
2019; 35: 653-62. doi: 10.1007/s10554-018-1482-2
9. Song L, Brezden-Masley C, Ramanan V, Ghugre N, Barfett JJ, Chan KK, et 
al. Serial measurements of left ventricular systolic and diastolic function 
by cardiac magnetic resonance imaging in patients with early stage breast 
cancer on trastuzumab. Am J Cardiol 2019; 123: 1173-9. doi: 10.1016/j.
amjcard.2018.12.046
10. Thavendiranathan P, Abdel-Qadir H, Fischer HD, Camacho X, Amir E, Austin 
PC, et al. Breast cancer therapy-related cardiac dysfunction in adult women 
treated in routine clinical practice: a population-based cohort study. J Clin 
Oncol 2016; 34: 2239-46. doi: 10.1200/jco.2015.65.1505
Radiol Oncol 2025; 59(4): 510-521.
Cheng SN et al. / Cardiotoxicity in cardiac magnetic resonance study patients, undergoing anti-HER2 therapy 521
11. Bouwer NI, Liesting C, Kofflard MJM, Brugts JJ, Kock MCJ, Kitzen J, et al. 
2D-echocardiography vs cardiac MRI strain: a prospective cohort study 
in patients with HER2-positive breast cancer undergoing trastuzumab. 
Cardiovasc Ultrasound 2021; 19: 35. doi: 10.1186/s12947-021-00266-x
12. Tarantini L, Cioffi G, Gori S, Tuccia F, Boccardi L, Bovelli D, et al. Trastuzumab 
adjuvant chemotherapy and cardiotoxicity in real-world women with breast 
cancer. J Card Fail 2012; 18: 113-9. doi: 10.1016/j.cardfail.2011.10.015
13. Esmaeilzadeh M, Urzua Fresno CM, Somerset E, Shalmon T, Amir E, Fan CS, 
et al. A combined echocardiography approach for the diagnosis of cancer 
therapy-related cardiac dysfunction in women with early-stage breast can-
cer. JAMA Cardiol 2022; 7: 330-40. doi: 10.1001/jamacardio.2021.5881
14. Curigliano G, Lenihan D, Fradley M, Ganatra S, Barac A, Blaes A, et al. 
Management of cardiac disease in cancer patients throughout oncological 
treatment: ESMO consensus recommendations. Ann Oncol 2020; 31: 171-
90. doi: 10.1016/j.annonc2019.10.023
15. Leong DP, Cosman T, Alhussein MM, Kumar Tyagi N, Karampatos S, 
Barron CC, et al. Safety of continuing trastuzumab despite mild cardio-
toxicity: a phase I trial. Cardio Oncology 2019; 1: 1-10. doi: 10.1016/j.jac-
cao.2019.06.004
16. Omland T, Heck SL, Gulati G. The role of cardioprotection in cancer 
therapy cardiotoxicity: JACC: CardioOncology state-of-the-art review. Cardio 
Oncology 2022; 4: 19-37. doi: 10.1016/j.jaccao.2022.01.101
17. Henry ML, Niu J, Zhang N, Giordano SH, Chavez-MacGregor M. 
Cardiotoxicity and cardiac monitoring among chemotherapy-treated breast 
cancer patients. JACC Cardiovasc Imaging 2018; 11: 1084-93. doi: 10.1016/j.
jcmg.2018.06.005 
18. Zhao M, Chen C, Zhang C, Xu X, Tian F, Wu B, et al. Cardiotoxicity with 
human epidermal growth factor receptor-2 inhibitors in breast cancer: 
disproportionality analysis of the FDA adverse event reporting system. Int J 
Cardiol 2023; 375: 87-93. doi: 10.1016/j.ijcard.2022.12.043 
19. Barbieri MA, Sorbara EE, Cicala G, Santoro V, Cutroneo PM, Franchina T, et 
al. Adverse drug reactions with HER2-positive breast cancer treatment: an 
analysis from the Italian Ph8-z.armacovigilance Database. Drugs Real World 
Outcomes 2021; 9: 91-107. doi: 10.1007/s40801-021-0027
20. Messroghli DR, Moon JC, Ferreira VM, Grosse-Wortmann L, He T, Kellman 
P, et al. Clinical recommendations for cardiovascular magnetic resonance 
mapping of T1, T2, T2* and extracellular volume: a consensus statement 
by the Society for Cardiovascular Magnetic Resonance (SCMR) endorsed by 
the European Association for Cardiovascular Imaging (EACVI). J Cardiovasc 
Magn Reson 2017; 19: 1-24. doi: 10.1186/s12968-017-0389-8
21. Haslbauer JD, Lindner S, Valbuena-Lopez S, Zainal H, Zhou H, D’Angelo T, et 
al. CMR imaging biosignature of cardiac involvement due to cancer-related 
treatment by T1 and T2 mapping. Int J Cardiol 2019; 275: 179-86. doi: 
10.1016/j.ijcard.2018.10.023
22. Jordan JH, D’Agostino RB, Jr., Hamilton CA, Vasu S, Hall ME, Kitzman DW, 
et al. Longitudinal assessment of concurrent changes in left ventricular 
ejection fraction and left ventricular myocardial tissue characteristics af-
ter administration of cardiotoxic chemotherapies using T1-weighted and 
T2-weighted cardiovascular magnetic resonance. Circ Cardiovasc Imaging 
2014; 7: 872-9. doi: 10.1161/CIRCIMAGING.114.002217