Breast cancer (BC) radiotherapy (RT) and resulting left ventricle (LV) exposure have been associated with subclinical LV dysfunction characterized by a global longitudinal strain (GLS) reduction>10-15% that could arise 6 months after RT. The acute phase of radiation response consists of inflammatory process. Fibrinogen, an essential coagulation factor which has also important role in inflammatory response, may be useful to enhance prediction of the risk of subclinical LV dysfunction in BC patients treated with RT.

To investigate the predictive value of pre-RT/post-RT plasma fibrinogen evolution in detecting patients with subclinical LV dysfunction defined as a GLS reduction>10% six months after BC RT.

The study consisted of BC patients treated with RT without chemotherapy. Speckle tracking echocardiography provided GLS before RT and 6 months postRT. Fibrinogen level was measured pre RT and at the end of RT and its evolution was defined by the ratio Post-RT-Pre-RT/Pre-RT (evoFIB). The prediction model was based on logistic regression including LV dose, evoFIB, and covariates. Area Under the Curve (AUC), sensitivity (Se), specificity (Spe) and misclassification error rate were considered in order to evaluate the predictive value of models.

The study included 75 BC patients. The addition of FIB, to LV dose and covariates, in the prediction model, allowed enhancing the AUC from 0.80 to 0.85 with an OR per 10% evoFIB increase of =1.60 (1.1–2.2). Se and Spe were increased from 67% to 75% and from 74% to 78% respectively. Information on evoFIB allowed decreasing misclassification error rate from 29% to 23%.

Collecting information on fibrinogen level pre and post RT in BC patients treated with RT could help to enhance prediction of subclinical LV dysfunction. The inflammatory process associated with coagulation in the acute phase post-RT may also be involved in mechanism of cardiac dysfunction observed at later phase.