Adoption of expansion margins to reduce the dose received by the coronary arteries and the risk of cardiovascular events in lymphoma patients

2020 
ABSTRACT Purpose Mediastinal radiotherapy (RT) in lymphoma patients implies involuntary coronary artery (CA) exposure, resulting in an increased risk of coronary artery disease (CAD). Accurate delineation of CAs may favor their sparing from higher RT doses. However, heart motion impacts on the estimation of the dose received by CAs. An expansion margin (PRV), encompassing the nearby area where CAs displace, may compensate these uncertainties, reducing CAs dose and CAD risk. Our study aimed to evaluate if a planning process optimized on CA-specific PRVs, rather than just on CAs, could provide any dosimetric and clinical benefit. Methods and materials Forty patients receiving RT for mediastinal lymphomas were included. We contoured left main trunk (LM), left anterior descending (LAD), left circumflex (CX) and right coronary (RCA) arteries. An isotropic PRV was then applied to all CAs, in accordance with literature data. A comparison was then performed by optimizing treatment plans either on CAs or on PRVs, to detect any difference in CAs sparing in terms of maximum (Dmax), median (Dmed) and mean (Dmean) dose. We then investigated if any dosimetric benefit obtained with PRV-related optimization process could translate in a lower risk of ischemic complications, through a risk modelling. Results Plan optimization on PRVs demonstrated a significant dose reduction (range 7-9%) in Dmax, Dmed and Dmean for the whole coronary tree, and even higher when vessels were located 5-to-20 mm far from PTV (range 13-15%), especially for LM and CX (range 16-21%). This translated in a mean risk reduction of developing CAD of 12% (p Conclusions Integration of CA-related PRVs in the optimization process reduces the dose received by CAs and translates in a meaningful prevention of the risk of CAD in lymphoma patients treated with mediastinal RT.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    32
    References
    2
    Citations
    NaN
    KQI
    []