ESTRO 2021 Abstract Book

S1267

ESTRO 2021

With the recent advances in deformable registration techniques, the ability of replacing control CTs (cCT) with CBCTs, as a trigger deciding about treatment plan adaptation for H&N cancer patients, was explored. Decreased cCTs number would decrease overall patient dose and improve the efficiency of the clinical workflow. Materials and Methods The planning CT (pCT) was deformably registered to the cCT and CBCT (same day acquisition), using a hybrid algorithm implemented in RayStation V8B (RaySearch Laboratories). Detailed visual and quantitative comparison between both registrations was performed. Additionally, original dose distribution was recalculated on the deformed images to evaluate differences in planning goals and dose statistics. Results Overall results between both registrations looked very promising. DVF comparison revealed similar absolute lengths (within 2mm) and no visual irregularities(fig1). Dose recomputation and dose statistics on the deformed images for important OARs and targets were comparable (within 1Gy). However, after the detailed inspection, 3 regions, where the algorithm was struggling to perform well, were identified: oral and nose cavities, metal artefacts and skull bones. In those areas local dose differences up to 8Gy(different cavity fillings), and up to 4Gy(artifacts) were observed(fig2). Conclusion For this patient CBCT would have triggered the same decision as cCT. However, when replacing a cCT with CBCT one needs to be careful. Systematic errors caused by the bone representation, metal and cavity fillings make algorithm very sensitive, which could cause errors in judgment. Finally, the presented workflow, is not entirely implemented in the TPS and therefore outsourcing to external software’s is still needed, which would affect the clinical workflow.

PO-1544 First experiences of online adaptive radiotherapy of vulvar cancer: Planned vs. realized dosimetry