ESTRO 2022 - Abstract Book

S295

Abstract book

ESTRO 2022

clinical implementation feasibility. In the future, a comparison between AI and human expert and a dosimetric evaluation will be performed.

PD-0331 To re-opt or not to re-opt: a study on the need for online plan adaptation for prostate cancer

L. Verweij 1 , S.U. Tetar 1 , O. Bohoudi 1 , B.J. Slotman 1 , A.M. Bruynzeel 1 , F.J. Lagerwaard 1 , M.A. Palacios 1

1 Amsterdam UMC, VUmc location, de Boelelaan 1117, 1081 HV, Department of Radiation Oncology, Amsterdam, The Netherlands Purpose or Objective Online adaptive MR-guided radiotherapy (MRgRT) involves re-contouring and plan re-optimization prior to each fraction to ensure adequate target coverage and maximal organ-at-risk sparing. Additionally, in general adaptive MRgRT is delivered using only minimal CTV to PTV safety margins. This online procedure is time-consuming and may not always be needed, e.g. when the interfractional anatomical changes are limited. In this study, we evaluated outcomes of target coverage and OAR doses when plan re-optimization was not performed using CTV to PTV margins of 3mm and 5 mm, respectively. Materials and Methods Ten patients with localized prostate cancer previously treated with MRgRT in 36.25 Gy in 5 fractions were included in this study. The CTV was partitioned in a CTV PR (prostate) and a CTV SV (seminal vesicles) by an experienced radiation oncologist. This was performed for all six MR-scans available during MRgRT, the simulation scan and high-resolution scans of fraction 1-5. All CTVs were expanded with a 3 and 5 mm margin, respectively, generating a PTV PR_3mm , PTV PR_5mm , PTV SV_3mm and PTV SV_5mm (Figure 1). Baseline plans were generated for both the 3- and 5mm margin plans. After alignment of the CTV, these baseline plans were recalculated for the anatomy of the day on each pre-treatment MRI (non-reoptimized plans). Target dosimetry (for both the 3- and 5mm plans) was evaluated by determining coverage of CTV PR and CTV SV by the 95% isodose line for each of the 50 fractions. A CTV coverage of V95% > 93% was considered as adequate. Doses to rectum and bladder were determined to evaluate the effect of applying varying PTV margins to relevant OARs.

Results The 3mm CTV to PTV margin plans had adequate coverage of the prostate and seminal vesicles in only 76% and 46%, respectively of all investigated 50 plans (Table 1). Increasing the CTV to PTV margins to 5mm ensured adequate coverage of the prostate in all but one fractions, however seminal vesicles coverage remained inadequate in 24% of plans. The improved coverage using the 5mm margins came at a cost of OAR doses. The bladder V 100% and V 90% increased on average 0.61cc and 3.33cc, respectively ( p<0.01 ). Similarly, the rectum V 100% and V 90% increased 0.25cc and 1.77cc, respectively ( p<0.01 ) in comparison to the 3mm plans.

Conclusion Without plan re-optimization and use of a 3 mm PTV margin, coverage of the CTV’s of the prostate and seminal vesicles was inadequate in about a quarter and half of the fractions, respectively. Increasing the margins to 5mm corrected prostate coverage at the expense of a higher OAR dose. Also with 5mm margins, coverage of seminal vesicles was inadequate in approximately one quarter of plans. These results underscore the need for routine online plan adaptation, particularly when using 3mm margins. Use of implanted gold markers in the prostate for CTV alignment on conventional linacs may be insufficient in intermediate- and high-risk patients with SV involvement.