2674 - Real-Life Impact of Hypofractionated IMRT in Early Breast Cancer on Cardiovascular Risk

A. A. Diaz-Gavela^1,2, E. del Cerro^1,2, F. Counago³, S. Sanchez⁴, C. Andreu¹, I. J. Thuissard¹, L. Guerrero⁴, M. Pena^1,2, V. Duque-Santana^1,2, Y. Molina⁵, and D. Sanz-Rosa¹; ¹Medicine Department, School of Biomedical Sciences, Universidad Europea, Madrid, Spain, ²Quironsalud Madrid University Hospital. Radiation Therapy Department, Madrid, Spain, ³San Francisco de Asís and La Milagrosa Hospitals. National Chair of Research. GenesisCare Spain, Madrid, Spain, ⁴La Luz Hospital. Radiation Therapy Department, Madrid, Spain, ⁵Hospital Quironsalud Madrid. Radiophysics Department, Madrid, Spain

Purpose/Objective(s): Various meta-analyses and studies based on historical registries have shown an excess of mortality and significant cardiovascular (CV) morbidity in breast cancer (BC) patients who receive adjuvant radiotherapy (RT) on the left side. However, these patients were treated with obsolete RT techniques, with conventional fractionation and without a precise knowledge of the heart dose-volume histograms (DVH) or CV-risk factors. We now know that the use of ultra-conformal techniques (IMRT) has led to a significant decrease in the heart dose, as reflected in the DVHs. In parallel, hypofractionated protocols -which result in lower EQD2 in organs at risk- have become the standard of care. The aim of our study was to assess whether the combination of both could reduce CV events compared to historical series and to identify clinical or dosimetric factors that could have a negative impact. Materials/

Methods: All BC patients at our institution who received hypofractionated (40.05Gy/15sessions) adjuvant forward-planned IMRT after conserving surgery were included. Patients in indication for elective lymph node RT and those who had received previous mediastinal RT were excluded.
Results: 882 patients with invasive BC or DCIS were analyzed (450 right breast; 432 left). After a median follow-up of 8ys (maximum 14), 35 CV events were identified (5 myocardial infarction/angina, 20 arrhythmias, 2 severe valvular disease and 7 "other non-ischemic”). No statistically significant differences were observed according to breast laterality (p=0.15). With regard to heart DVHs, none of those analyzed had a deleterious effect, although there was a slight and progressive trend towards statistical significance (Table). Patient-dependent CV risk factors that were found to have a significant impact on the incidence of late CV events were dyslipidemia, hypertension, family history of heart disease and overweight/obesity (p= 0.03, <0.001, <0.001 and 0,02 respectively).
Conclusion: Although the deleterious impact of thoracic RT on CV late events is well known, in our cohort it has been shown to be less determinant than patient-dependent baseline risk factors for any CV event. Further long-term follow-up will be necessary to determine whether these results are sustained over time and, if event curves continue to diverge, to assess the extent to which they do so. On the other hand, a precise understanding of the impact of CV risk factors may help to design targeted primary prevention strategies from which BC patients may benefit.

Long term cardiac toxicity
Heart DVH (median [ICR])	Yes (n=35)	No (n=847)	p-value
V161	0 [0.28]	0 [1.12]	0.88
V8 1	0 [0.64]	0 [1.72]	0.086
V1 1	4,32 [12.49]	7.48 [13.65]	0.070
V0.51	11.93 [17.54]	17.30 [18.22]	0.069
Dmax*	4.08 [33.4]	7.12 [35.32]	0.054
D2*	2,12 [5.20]	2.52 [20]	0.128
Dmed*	0.60 [1.32]	0.68 [1.64]	0,391
1 % of the heart who receives 16, 8, 1 and 0.5Gy respectively; * Expressed in Gy; DVH: dose-volume histograms; ICR: Interquartile range. Dmax: Maximum heart dose; Dmed: Mean heart dose.