ESTRO 2021 Abstract Book

S499

ESTRO 2021

1 Institut Jules Bordet - Université Libre de Bruxelles, Medical Physics Department, Brussels, Belgium; 2 Medizinische Universität Wien, Center for Medical Physics and Biomedical Engineering, Vienna, Austria; 3 Institut Jules Bordet - Université Libre de Bruxelles, Radiation Oncology, Brussels, Belgium Purpose or Objective Immunotherapy has created an increasing interest in exploiting the immune system and its interaction with radiotherapy (RT), as well as its influence on the outcome of cancer patients. Even though RT is one of the contributing causes of lymphopenia (LP), lymphocyte-rich organs at risk (LOAR) are rarely used to optimise and evaluate RT treatment plans. The purpose of this work is to evaluate the feasibility of including LOAR in plan optimisation, while maintaining adequate target coverage and dose constraints to the classical OARs (cOARs) for stage III non-small cell lung cancer (NSCLC). Materials and Methods Twenty randomly selected stage III NSCLC patients treated with a curative intent normo-fractionated RT scheme were included. In addition to the nodal target volume (PTVn), 16 patients presented a primary tumour at treatment time, defining a total target volume (PTVt). For nodal irradiation, both the involved site (ISRT) and the involved node (INRT) approaches were considered. In addition to the cOARs, thoracic vertebrae bodies (TVBs) and great vessels (GVs) were contoured and considered as surrogates for the haematopoietic and circulatory blood cells, respectively. The impact of the lymphocyte-sparing (LS) approach was evaluated on both ISRT and INRT delineations. For each, two VMAT plans were generated with the Monaco TPS (v5.51, Elekta AB, Stockholm, Sweden): one standard-of-care (SOC) plan optimising solely on the cOARs, and one LS plan adding the LOARs. 66Gy/33# was prescribed to the D50% of either the PTVt or PTVn (no D50% inferior to the prescribed dose was accepted). For the LS plans, no violation of the mandatory dose constraints for the cOARs was accepted. Differences between both plans pairs in mean dose of heart, lungs, TVB, GV and oesophagus, as well as lung V5Gy, V20Gy, V30Gy and the spinal canal D0.1cc were assessed using a Wilcoxon signed-rank test or a paired t-test and considered statistically significant if p<0.05. Results The results are summarised in table 1.

While there are statistically significant differences for the V90% and D2% (IS vs LSIS), V95% (IN vs LSIN), all plans respect the V95% objective of the PTVs. In the LSIS, there is an increase in all the dosimetric parameters of the lungs. While statistically significant, this increase remains clinically acceptable. No difference was found for LSIN, nor for the heart. All the dosimetric parameters differences evaluated for the LOARs were statistically significant, with a greater decrease on the TVBs for the LSIS vs LSIN. Regarding the GVs, the mean dose decrease between SOC and LS plans is greater with the LSIN (figure 1). The reduction of the dose to the spinal cord is