ESTRO 2024 - Abstract Book

S4138

Physics - Intra-fraction motion management and real-time adaptive radiotherapy

ESTRO 2024

study, we selected ten fractions, from eight patients, with PGI-based detection of relevant treatment deviations that were also retrospectively confirmed by anatomical changes in the respective cCTs. These changes originated from variations in subcutaneous tissue, femur rotations and/or organ fillings. All patients had robustly optimized initial simultaneous integrated boost plans, with two lateral opposing fields delivering 48Gy/60Gy to the low-risk/high-risk clinical target volume (CTV Low /CTV High ) in 20 fractions. In addition to the initial plan, three scenarios were investigated on the cCTs: the non-adapted, the partially adapted and the fully adapted plan. To retrieve the non-adapted dose distributions, the initial plans were re-calculated on the cCTs. The optimal fraction dose distributions were obtained by fully re-optimized plans on the cCTs. Partially adapted fraction doses considered the non-adapted field for which PGI had detected the relevant treatment deviation as background dose. A script-based partial adaptation was implemented in RayStation (v.11.0.100), performing robust mimicking of a “reference dose” (fully adapted dose minus background dose) for the remaining field [2]. The resulting dose distribution was summed with the background dose to obtain the partially adapted fraction dose.

A dosimetric assessment compared the four plans by analysing CTV coverage ( D 98% >95%), CTV hotspots ( D 2% <110%) and organ at risk (OAR) sparing.

Results:

In all ten fractions (Fig.2), both partial and full adaptation led similarly to sufficient coverage of the CTVs (median D 98% for CTV Low /CTV High : 100.7%/98.0% for partial adaptation and 100.6%/98.6% for full adaptation), while the D 98% of non adapted plans was lower, especially for the CTV High , with median values of 98.8% and 95.7% for CTV Low and CTV High , respectively. Median D 98% values for the initial plans were 99.4%/98.6% (CTV Low /CTV High ). Hotspots dose D 2% to the CTV High was always below clinical constraints with median values of 101.3%, 101.4%, 102.2% and 101.6% for the initial plan, no, partial and full adaptation, respectively. Dose values in OARs were case dependent for the initial plans. Dosimetric changes in rectum ( V 57Gy ) and penile bulb ( V 49Gy ) were similarly smaller for both the partially and the fully adapted plans compared to the non-adapted plans, which led more often to dose parameters above the clinical thresholds. For OARs closer to the CTVs, like the bladder, an increase in the investigated dosimetric parameter ( V 60Gy ) was observed for partial adaptation in comparison with the other plans, however, all results were within constraints.