Functional CMR Metrics for Risk Assessment in Childhood Cancer Survivors: Insights from MyoHealth and GLS

Saturday, February 7, 2026

10:36 AM - 10:43 AM

Location: Galapagos / Aruba / Ilhabela (2nd Lower Level)

Presenting Author(s)

Olga H. Toro-Salazar, MD

Head of Dimensional Imaging and Cardio-Oncology
Vanderbilt University Medical Center
Nashville, Tennessee, United States

Co-Author(s)

RL

Raymond P. Lorenzoni, MD

Noninvasive Imaging and Outreach
Connecticut Children's
Hartford, Connecticut, United States
TF

Tongyao Fan, PhD, MPH

Research Scientist
Connecticut Children's Medical Center, United States
TB

Tiffany H. Berthod, RN

Cardio-Oncology Nurse Coordinator and Analyst
Connecticut Children's
Hartford, Connecticut, United States
DB

David Bearl, MD, MA

Assistant Professor, and Medical Director of the Ventricular Assist Device Program
Vanderbilt University Medical Center, United States
Jonathan H. Soslow, MD, MSc

Professor, Pediatrics, Division of Pediatric Cardioloty
Vanderbilt University Medical Center
Nashville, Tennessee, United States
DP

David Parra, MD

Associate Professor
Vanderbilt University Medical Center, United States
Sudeep Sunthankar, MD

Physician
Vanderbilt University Medical Center
Nashville, Tennessee, United States
MD

Matthew Dove, MD

Pediatric Cardiologist
Vanderbilt
Jackson, Tennessee, United States

Primary Author(s)

Olga H. Toro-Salazar, MD

Head of Dimensional Imaging and Cardio-Oncology
Vanderbilt University Medical Center
Nashville, Tennessee, United States

Background: Childhood cancer survivors (CCS) are at increased risk for cardiovascular (CV) disease due to prior exposure to cancer therapy. Quantitative cardiac magnetic resonance (CMR)-derived MyoHealth (Figure 1a) provides a novel approach to assess subclinical myocardial dysfunction, but its relationship to traditional CV and cancer-related risk factors remains incompletely understood.

Methods: From April 2022 to July 2025, 70 unique CCS underwent 90 CMR studies with available CMR-derived MyoHealth scores. Patients were stratified into 3 MyoHealth categories: normal (≥70%), borderline (55-70%), and abnormal (≤55%). Stratification was chosen based on adult studies demonstrating a higher risk of adverse outcomes with MyoHealth ≤55%. Demographics, cancer treatment-related, and cardiometabolic variables were compared across groups; differences in cardio-oncology risk factors were examined between normal, borderline and abnormal groups. A secondary classification incorporating both MyoHealth and left ventricular global longitudinal strain (LV GLS) was also explored. Continues variables were compared with Kruskal-Wallis test, and categorical variables were compared with chi-square or Fisher's exact test.

Results: The median age at first CMR was 18.5 years; 48% were female. Based on the Myohealth score at the first CMR, 23/70 (33%) CCS were classified as normal, 22/70 (31%) borderline, and 25/70 (36%) abnormal (Table 1). CCS with abnormal MyoHealth demonstrated higher BMI (median 28.7 vs. 21.3 kg/m²) and heart rate despite lower cumulative AC doses (162.7 vs. 313.5 mg/m²) compared to normal group. Among the 90 CMR studies, CCS with borderline (n=31) and abnormal (n=30) MyoHealth had higher CV risk scores (median 3.0 and 4.0 vs. 2.5, p=0.02) and greater prevalence of hypertension (32.3% and 23.4% vs. 13.7%, p=0.006) as compared with normal (n=29) (Table 2). CCS with borderline and abnormal MyoHealth also had a trend towards higher rates of being overweight/obese (41.9% and 60.0% vs. 27.6%, p=0.04) and prior cancer therapy-related cardiac dysfunction (CTRCD) (96.8% and 96.7% vs. 79.3%, p=0.03).

Conclusion: CMR-derived MyoHealth effectively stratifies CV risk in CCS. Those with abnormal values demonstrated more frequent obesity, hypertension, and CTRCD despite less AC exposure and overall lower cancer risk scores. These findings suggest that functional CMR metrics may enhance long-term risk assessment beyond traditional treatment exposure, highlighting the importance of personalized survivorship care. Borderline MyoHealth scores (55-70%) represent clinically significant findings that require appropriate evaluation and management.

Table 1. Demographics and Baseline MyoHealth Score (n=70)

	Total participants n=70	Normal (MyoHealth≥70%) n=23	Borderline (MyoHealth 55-70%) n=22	Abnormal (MyoHealth≤55%) n=25
Demographics	Median (IQR)
Current age (in years) *	18.5 (7.0)	18.0 (7.5)	19.5 (7.0)	18.0 (6.0)
Age at Cancer Diagnosis (in years)	13.0 (10.0)	14.0 (12.0)	12.0 (9.5)	14.0 (10.0)
Height (cm) *	164.6 (16.3)	162.0 (16.7)	167.1 (12.6)	163 (15.5)
Weight (kg) *	69.5 (35.1)	60.4 (27.8)	71.1 (28.3)	77.5 (49.7)
BSA *	1.8 (0.5)	1.7 (0.5)	1.8 (0.5)	1.9 (0.6)
BMI *	23.5 (9.9)	21.3 (7.2)	23.5 (5.4)	28.7 (15.0)
HR (bpm) *	74.0 (19.5)	70.0 (18.5)	74.0 (14.8)	79.0 (17.0)
sBP (mm Hg) *	110.0 (16.5)	108.0 (21.0)	115.5 (12.8)	106.0 (16.0)
dBP (mm Hg) *	59.0 914.0)	58.0 (13.5)	63.0 (14.0)	59.0 (15.0)
Total cumulative anthracyclines dose (mg/m²)	245.4 (222.5)	313.5 (213.0)	314.8 (174.8)	162.7 (132.6)
Total radiation dosage, Gy	22.5 (21.0)	42.8 (26.5)	19.0 (9.6)	21.0 (22.2)
	N (%)
Female	34 (48.4)	11 (47.8)	13 (59.1)	10 (40.0)
Ethnicity: Hispanic and/or Latino	23 (32.9)	7 (30.4)	5 (22.7)	11 (44.0)
Race, white	41 (58.6)	11 (47.8)	16 (72.7)	14 (56.0)
Cancer Diagnosis (5 subjects had dual cancer diagnoses)
Lymphoma	20 (28.2)	6 (26.1)	3 (13.6)	10 (40.0)
Leukemia	22 (31.0)	5 (21.7)	9 (40.9)	8 (32.0)
Solid Tumor	30 (42.3)	11 (47.8)	12 (54.5)	7 (28.0)
Other	4 (5.6)	2 (8.7)	0 (0)	2 (8.0)
On active cancer therapy when CMR completed*	15 (21.4)	6 (26.1)	2 (9.1)	7 (28.0)
>1 CMR performed	15 (21.1)	3 (13.0)	5 (22.7)	7 (28.0)

*At the time of first CMR

BMI=Body Mass Index; BSA=Body Surface Area; CMR=Cardiovascular Magnetic Resonance; DBP= Diastolic Blood Pressure; Gy=Gray; HR=Heart Rate; IQR= Interquartile Range; SBP=Systolic Blood Pressure.Table 2. Cardiovascular and Cancer Risk Factors in MyoHealth Groups Among 90 Consecutive CMR Studies

Risk factors		Normal (Myohealth≥70%) n=29	Borderline (Myohealth 55-70%) n=31	Abnormal (Myohealth≤55%) n=30	P-value*
Total cardio-oncology risk score, median (IQR)		9.5 (4.5)	9.5 (4.0)	8.0 (1.5)	0.36
Total cancer risk score, median (IQR)		6.0 (3.5)	6.5 (3.8)	3.8 (3.3)	0.008
Total cardiovascular risk score, median (IQR)		2.5 (2.0)	3.0 (1.0)	4.0 (1.5)	0.02
	Cardiovascular related risk categories
Previous CTRCD, n (%)		23 (79.3)	30 (96.8)	29 (96.7)	0.03
BMI, n (%)					0.06
< 85th percentile or BMI 18.5-24.9		21 (72.4)	18 (58.1)	12 (40.0)
85th-95th percentile or BMI 25-29.9		5 (17.2)	8 (25.8)	5 (16.7)
≥ to 95th percentile or BMI ≥ 30		1 (3.4)	4 (12.9)	7 (23.3)
≥ 120% of the 95th percentile OR BMI ≥ to 35 (whichever is lower based on age and sex)		2 (6.9)	1 (3.2)	6 (20.0)
BMI: Overweight/obese, n (%)		8 (27.6)	13 (41.9)	18 (60.0)	0.04
Lipids, n (%)					0.59
LDL-c < 110 mg/dL, non-HDL-c < 120 mg/dL, AND triglycerides < 150 mg/dL)		14 (48.3)	19 (61.3)	15 (50.0)
LDL-c 110-129 mg/dL, OR non-HDL-c 120-144 mg/dL, OR triglycerides 150-199 mg/dL		3 (10.3)	5 (16.1)	6 (20.0)
LDL-c ≥ 130 mg/dL, OR non-HDL-c ≥ 145 mg/dL, OR triglycerides ≥ 200 mg/dL		5 (17.2)	5 (16.1)	4 (13.3)
Unknown or unspecified		7 (24.1)	2 (6.5)	5 (16.7)
Hypertension, n (%) **					0.006
Normal		25 (86.2)	21 (67.7)	23 (76.6)
Elevated		3 (10.3)	7 (22.6)	0 (0)
Stage 1		0 (0)	3 (9.7)	5 (16.7)
Stage 2		1 (3.4)	0 (0)	2 (6.7)
Diabetes, n (%)					0.08
Normal		19 (65.5)	23 (74.2)	15 (50.0)
Prediabetes		0 (0)	3 (9.7)	3 (10.0)
Diabetes		0 (0)	0 (0)	2 (6.7)
Unknown		10 (34.5)	5 (16.1)	10 (33.3)
Ferritin, n (%)					0.40
< 1000 µg/L		14 (48.3)	21 (67.7)	20 (66.7)
>1000 µg/L		9 (31.0)	6 (19.4)	8 (26.7)
Unknown		6 (20.7)	4 (12.9)	2 (6.7)
Cardiorespiratory fitness, n (%)					0.75
Good-Superior cardiorespiratory fitness based on relative VO₂ max for age & sex ( > 80% of predicted value)		5 (17.2)	6 (19.4)	3 (10.0)
Fair-Very Poor cardiorespiratory fitness based on relative VO₂ max for age & sex (< 80% of predicted value)		5 (17.2)	9 (29.0)	7 (23.3)
Functional disability based on relative VO₂ max for age & sex		4 (13.8)	3 (9.7)	2 (6.7)
Unknown		15 (51.7)	13 (41.9)	18 (60.0)
Cancer related risk categories
Age at cancer diagnosis, n (%)					0.53
>=5 years		22 (75.9)	24 (77.4)	25 (83.3)
1-4 years		5 (17.2)	7 (22.6)	5 (16.7)
< 1 year		2 (6.9)	0 (0)	0 (0)
Female, n (%)		12 (41.4)	16 (51.6)	12 (40.0)	0.61
Radiation: to heart region only, n (%)					0.008
None		21 (72.4)	17 (54.8)	22 (73.3)
< 5 Gy		0 (0)	0 (0)	1 (3.3)
5-15 Gy		0 (0)	4 (12.9)	2 (6.7)
15-30 Gy		0 (0)	7 (22.6)	2 (6.7)
>30 Gy		8 (27.6)	3 (9.7)	3 (10.0)
Transplant, n (%)		7 (24.1)	2 (6.5)	3 (10.0)	0.03
Vinca alkaloids use, n (%) ***		18 (62.1)	24 (77.4)	26 (86.7)	0.10
Alkylating agents use, n (%)		26 (89.7)	26 (83.9)	28 (93.3)	0.54
Cumulative anthracyclines dose, n (%)					0.02
< 101 mg/m²		6 (20.7)	2 (6.5)	8 (26.7)
101-200 mg/m²		5 (17.2)	8 (25.8)	9 (30.0)
>200-250 mg/m²		1 (3.5)	3 (9.7)	6 (20.0)
>250-300 mg/m²		0 (0)	0 (0)	0 (0)
>300 mg/m²		17 (58.6)	18 (58.1)	7 (23.3)

* Medians were compared using the Kruskal-Wallis test; categorical variables were compared using chi-square or Fisher’s exact test

**Hypertension status was obtained from the CMR report

***Only count when given in combination with anthracycline

BMI=Body Mass Index; CTRCD=Cancer Therapy-Related Cardiac Dysfunction; Gy=Gray; LDL-c=Low-Density Lipoprotein Cholesterol; IQR= Interquartile Range; VO₂ max=maximal oxygen consumption.Figure 1 is composed of two panels, 1a and 1b, that illustrate a cardiovascular magnetic resonance (CMR) imaging technique and a risk stratification scoring pathway.