Heart dose variability in RT of left breast cancer patients treated in deep inspiration breath hold
Gracinda Johansson,
Sweden
PO-1480
Abstract
Heart dose variability in RT of left breast cancer patients treated in deep inspiration breath hold
Authors: Gracinda Johansson1, Emil Fredén1, Johan Knutsson1, Martin Olin1, Linda Dagertun1, Albert Siegbahn1,2
1Södersjukhuset, Department of Oncology, Stockholm, Sweden; 2Karolinska Institutet, Department of Clinical Sciences and Education, Stockholm, Sweden
Show Affiliations
Hide Affiliations
Purpose or Objective
To evaluate the impact of the variability of the heart
position and shape in the heart dose for patients with left-sided breast cancer
who received external beam radiotherapy (RT) in deep inspiration breath hold
(DIBH).
Material and Methods
A total of 150
treatment fractions for 10 patients who received RT of left-sided breast cancer
(2.67 Gy x 15 fx) with the DIBH technique were included in this study. The
patients were initially positioned with the Catalyst (C-RAD, Uppsala, Sweden)
surface scanning system in free breathing position. A CBCT scan was
subsequently acquired in treatment position (DIBH) to derive the necessary
couch translations to align the patient in accordance with the planning CT
(pCT). The daily CBCT images were exported
to the Monaco (Elekta AB, Stockholm, Sweden) treatment planning system (TPS).
The online couch translations were thereafter used to rigidly register the CBCT
images to the pCT. The pCT was
deformably registered to all the daily CBCT images (using the
‘Adapt-To-Anatomy’ functionality in Monaco), and the reference heart contour
was subsequently propagated to each CBCT to generate 15 ‘heart-of-the-day’
contours. The daily heart contours were then rigidly copied to the pCT. Thus,
for each patient we had 16 heart contours segmented in the pCT: 1 reference heart contour and 15 representing
the day-to-day variation in heart position and shape (Figure 1). To estimate
the heart dose for each fraction, we re-calculated the treatment plan 15
times. For each re-calculation, a
relative electron density (ED) of 1 (relative to water) was assigned to the heart-of-the-day
contour and for the reference heart contour, the relative ED was set to 0.01. The
body contour of the pCT was assumed to be representative for the geometry of
each fraction. For each treatment fraction, the mean heart dose was registered and
the differential DVH of the heart was exported and used as input in the
calculation of the NTCP (risk for late cardiac morbidity), using the relative
seriality model.
Results
For all patients, the mean heart dose calculated on
the pCT did not exceed our planning constraint of 4.005 Gy. The mean heart dose
calculated for the different treatment fractions, using the heart contour
transferred from the daily CBCT, ranged from -21.1 % to 58.7 %, relative to the
mean heart dose determined for the reference heart contour. A summary
of the results is presented in Figure 2. The NTCP for late cardiac
morbidity for the reference heart was 0 % for all patients. However, for the heart contour
taken from the CBCT images, the NTCP varied between 0 % and 0.33 %.
Conclusion
The variation in
the heart position and shape between fractions could lead to higher doses being
delivered to the heart, compared to the dose calculated on the pCT. This could
lead to an underestimation of the expected heart NTCP due to RT. These
variations should be considered in the plan evaluation for left-sided breast
cancer patients receiving RT in DIBH.