Click here to


Are you sure ?

Yes, do it No, cancel

Cardiotoxicity Mitigation in Central Lung Hypofractionated Radiotherapy Using Non-Uniform Spatiotemporal Dose Modulation

A Modiri1*, A Mendes2 , S Lam2 , A Berman2 , S Feigenberg2 , A Sawant1 , (1) University of Maryland, School of Medicine, Baltimore, MD, (2) University of Pennsylvania, Philadelphia, PA


(Monday, 7/15/2019) 3:45 PM - 4:15 PM

Room: Exhibit Hall | Forum 4

Purpose: Incidence of radiation-induced major coronary events in thoracic cancer patients continues to adversely impact patient survival with a reported 7.4% increase per every 1Gy of mean heart dose. This fact suggests that there is no “safe� dose threshold for the normal heart tissue. To minimize the cumulative heart dose over treatment fractions, we developed and investigated a spatiotemporally-optimized radiotherapy planning technique (SORPT) where both inter-fraction beam-angles and dose distributions were considered as variables in the treatment plan optimization process.

Methods: We retrospectively analyzed data from a stereotactic body radiation therapy (SBRT) patient with central lung tumor who was treated with an 8-beam conformal radiotherapy clinical plan (CLP) where dose to 95% of the planning target volume (93cc) was 60Gy (3-fx) and mean/max heart doses were 8/34Gy. In our in-house SORPT implementation, the inverse plan objective function consisted of maintaining tumor coverage, minimizing dose to the heart and meeting commonly-used dose-volume constraints for other organs. To allow summation of heterogenous fractional voxel doses, we used biologically effective dose in formulations. While the number of fractions was kept equal to that of the CLP, for each fraction, SORPT chose from a pool of 24 beam-angles, including CLP beam-angles and their ±10-degree shifted versions, and optimized dose distribution. Single-fraction dose constraints were applied to avoid excessive inter-fraction dose heterogeneity.

Results: The SORPT chose a set of [6, 6, 7] beams for fractions [1, 2, 3] and created a non-uniform dose distribution across fractions. Overall, 15 distinct gantry-angles were chosen, 20% of which were common with CLP. The overall mean/max heart doses were reduced by 12.5/11.8 % compared to the CLP. Doses to spinal cord, esophagus and normal lung stayed below their known thresholds.

Conclusion: We demonstrated the potential of employing spatiotemporal dose heterogeneity for reducing post-treatment cardiotoxicity in thoracic cancer survivors.


Treatment Planning, Heart


TH- External beam- photons: General (most aspects)

Contact Email