ESTRO 2024 - Abstract Book

S3984

Physics - Inter-fraction motion management and offline adaptive radiotherapy

ESTRO 2024

During the day-night cycle the height of the human body changes. The spine length decreases during the day due to fluid loss and deformation of the discs, caused by gravity and applied stress to the body mass 1 . This change in height of the human body, called diurnal spine length variation, is a well-known phenomenon, but has not yet been quantified in the field of radiotherapy. If a planning CT (pCT) is acquired at a time of day different from the daily CBCT imaging during treatment, a systematic difference might be induced due to this difference in spine length. Therefore, the aim of this study was to quantify the spine length variation in pediatric patients undergoing radiotherapy, and to investigate the correlation of the time interval between image acquisitions with spine length variation, as well as the correlations of patient’s height and weight with spine length variation.

Material/Methods:

For this study, we included twenty pediatric cancer patients (mean age 10.3; range 3.3–16.0 years), who had been treated between 2012-2018 with image guided radiotherapy in two institutes (Table 1). Patients were selected when they had two fractions per day, or fractions were planned on varying time slots over the course of treatment, during which abdominal CBCTs were acquired (100–120 kV, 10–40 mA, gantry rotation 195–360 degrees). A total of 20 pre treatment pCTs (120 kV, slice thickness 2.0–7.5 mm), and 104 CBCTs images had been acquired; 52 in the morning and 52 in the afternoon. The CBCTs were registered to the pCTs using two automatic registrations relative to the bony anatomy; one relative to the T11 and one to the L4 vertebra. For each CBCT, the difference between the cranial-caudal (CC) displacement of T11 and the CC displacement of L4 was calculated. To determine the diurnal spine length variation (ΔLD), the difference in CC displacements between the morning and afternoon CBCTs was calculated. For all 52 pairs of morning and afternoon CBCTs, the mean ±SD and ranges of the spine length variation (ΔLD) were calculated, and tested for significance (non-parametric Sign Test, significance value p<0.05). Furthermore, using the Spearman’s ρ (significance value p<0.05), we investigated the correlation of the interval time (in hours) between the two image acquisitions and the spine length variation, and the correlations of patients’ height and weight with the spine length variation.