ESTRO 2021 Abstract Book

S1336

ESTRO 2021

70/60/54Gy. NAL protocol was performed acquiring orthogonal KV-MV images the five first days and once per week and CBCT was also acquired weekly. Shoulder structures were contoured in TPS in order to avoid the beam entering through them, in 44 patient treatments (group A). For 53 patients (group B) 6mm margin was given to the shoulder structures. Transit in vivo dosimetry with EPID was performed in the first fraction and every week. A number of 699 fractions have been analyzed automatically using PerFraction TM (Sun Nuclear) with local gamma criteria 3%,3mm and 5%,5mm, 20% threshold and 95% tolerance to the predicted dose distribution. On the same fractions a 3D log-based dose calculation in the CBCT was also performed using PerFraction TM . Comparison of the gamma passing rate (%GP) between two groups of patients (A and B) was performed to assess the plan robustness to shoulders position. We hypothesize that while the shoulder position would give random differences, loss of weight (i.e. diameter change) would give a decline tendency in %GP. Therefore, for patients showing a decrease of %GP, the difference of diameter between the first and last session was assessed to detect any false positives. In order to detect false negatives, changes of diameter were checked in CBCTs for a group of patients not showing a decline in %GP. The dosimetric impact in the CTVi was also determined by comparing 3D %GP recalculating dose on CBCT (<95% of points gamma 3% local dose-3mm). Results The 2D gamma analysis of group A and B are shown in figure 1. No significant differences were found (p > 0.05). From the total number of patients studied, in 17 of them a decline in gamma passing rate was detected (figure 2). From those, 13 had a reduction in diameter between 1 and 2 cm. This had a dosimetric impact in the recalculation on the CBCT for 5 patients. For the rest of the patients for which no decline tendency was detected, only in 5 cases PerFraction 2D transit evaluation failed to detect the diameter reduction.

Conclusion Transit in vivo dosimetry allows to detect weight loss in H&N patients by assessing gamma passing rate tendency along treatment. The sensitivity for diameter changes is 72% and specificity 95%. Strategies for H&N treatment plans robustness to shoulders position showed small, non-significant, differences in gamma passing rate for the 2D analysis. Image segmentation focusing on shoulders would improve this evaluation. PO-1615 Pre-treatment verification of SBRT using novel OCTAVIUS detector 1600SRS and portal dosimetry. B. Bajon 1 , M. Kruszyna-Mochalska 1 , P. Romański 1 1 Greater Poland Cancer Centre, Medical Physics, Poznań, Poland

Purpose or Objective The purpose of the study was to implement the new 2D-array 1600SRS with OCTAVIUS rotational phantom for