AUTHOR=Locquet Médéa , Spoor Daan , Crijns Anne , van der Harst Pim , Eraso Arantxa , Guedea Ferran , Fiuza Manuela , Santos Susana Constantino Rosa , Combs Stephanie , Borm Kai , Mousseaux Elie , Gencer Umit , Frija Guy , Cardis Elisabeth , Langendijk Hans , Jacob Sophie 

TITLE=Subclinical Left Ventricular Dysfunction Detected by Speckle-Tracking Echocardiography in Breast Cancer Patients Treated With Radiation Therapy: A Six-Month Follow-Up Analysis (MEDIRAD EARLY‐HEART study)

JOURNAL=Frontiers in Oncology

VOLUME=Volume 12 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2022.883679

DOI=10.3389/fonc.2022.883679

ISSN=2234-943X

ABSTRACT=Background: In case of breast cancer (BC), radiotherapy (RT) allows reducing locoregional recurrence and BC-related deaths but can lead to cardiotoxicity, resulting in an increased risk of long-term major cardiovascular events. It is therefore of primary importance to early detect subclinical left ventricular (LV) dysfunction in BC patients after RT and to determine the dose-response relationships between cardiac doses and these events.
Methods: Within the frame of the MEDIRAD European project (2017-2022), the prospective multicenter EARLY HEART study (ClinicalTrials.gov Identifier: NCT03297346) included chemotherapy naïve BC women aged 40-75 years and treated with lumpectomy and adjuvant RT. Myocardial strain analysis was provided using speckle tracking echocardiography performed at baseline and 6 months following RT. A global longitudinal strain (GLS) reduction >15% between baseline and follow-up was defined as a GLS-based subclinical LV dysfunction. Individual patient dose distributions were obtained using multi-atlas-based auto-segmentation of the heart. Dose-volume parameters were studied for whole heart (WH) and left ventricle (LV). 
Results: The sample included 186 BC women (57.5±7.9 years, 64% left-sided BC). GLS-based subclinical LV dysfunction was observed in 22 patients (14.4%). These patients had significantly higher cardiac exposure regarding WH and LV doses compared to patients without LV dysfunction (for mean WH dose: 2.66±1.75 Gy versus 1.64±0.96 Gy, p = 0.01). A significant increased risk of subclinical LV dysfunction was observed with the increase in the dose received to the WH (ORs from 1.13 (V5) to 1.74 (Dmean); p <0.01) and to the LV (ORs from 1.10 (V5) to 1.46 (Dmean); p <0.01). Based on ROC analysis, LV-V5 parameter may be the best predictor of the short-term onset of subclinical LV dysfunction. 
Conclusion: These results highlighted that all cardiac doses were strongly associated with the occurrence of subclinical LV dysfunction arising 6 months after BC RT. Whether measurements of GLS at baseline and 6 months after RT combined with cardiac doses can early predict efficiently subclinical events occurring 24 months after RT remains to be investigated.