ESTRO 2020 Abstract book

S244 ESTRO 2020

Ten lung cancer patients treated from 2017 and 2019 with a Histological diagnosis of NSCLC or SCLC were identified. 4D CT acquisition were performed, iCTV-PTV margin of 6 mm in cranial-caudal direction and 4 mm in radial direction were applied. Dose prescription were 60 Gy in 30 fraction, PTV coverage prescription was D95%≥95%. IMRT versus V-MAT RT plan were generated. Daily online CBCT images were acquired. We retrospectively recontured GTV_T, CTV and Organs at risk (OAR) for all 30 CBCT for all patients (total 300 CBCT). A DVH representing each single fraction and a mean DVH representing all 30 fractions were generated using an in-house software (PRO- DVH). Mean plan_adaptive was compared to plan_reference. Z test was exploited to detect statistically significant differences between provisional and adaptive CVT coverage. A shrinkage GTV_T map was also generated. Results In eight of 10 patients, mean target coverage of plan_adaptive resulted equal or superior to plan_reference. In two cases CTV coverage resulted inferior to plan_reference (D95=90% versus D95=95% and D95%=87,5% versus D95=95%, respectively, all p<0,001). In one patient, the analysis of DVH_adaptive of each fraction showed a progressive reduction in target coverage in the last 10 fractions. No trends were identified for the second patient. Concerning GTV_T volumes, we observed a mean regression in the whole series of 48% (range 32%-65%), with a major trend of reduction between the VIII and the XX fractions (figure 1).

how RILD evolves during the 24-months after RT using CT imaging and breathing function tests. Material and Methods CT images were available pre-RT and at 3, 6, 12 and 24- months following RT for forty-five subjects enrolled in IDEAL-CRT; a stage I/II clinical trial of isotoxic, dose- escalated radiotherapy for locally advanced non-small cell lung cancer. Patients were treated with 5 or 6 fractions a week and all received concurrent chemotherapy. To investigate how RILD appears and develops, radiological findings were quantified using a suite of CT-based imaging biomarkers (previously developed by our group) over serial time-points. The biomarkers are representative of parenchymal, pleural and lung volume change, and anatomical distortions after RT. Medical Research Council (MRC) dyspnoea scores and formal pulmonary function tests (PFTs) were also available for the majority of the subjects included. Results Radiological findings and breathing function vary according to time since RT. Figure 1 shows temporal change of selected biomarkers and breathing function. Parenchymal damage was common at 3-months and peaked at 6-months. From 6 to 24-months, the volume of consolidated parenchyma reduced and the normal lung volume in the treated lung progressively reduced. Anatomical distortions increased over time, peaking at 12 to 24-months. Pleural change was common at all time- points and demonstrated variable evolution across the cohort. FEV 1 /FVC % and MRC scores progressively declined after RT, with 48% of the subjects reporting increased breathlessness at 24-months. Based on radiological findings, we recognised two patterns of RILD evolution: patients whose normal lung volume shrinkage peaked at 6 months and then improved, indicative of abating acute inflammation (Group A, n=24), and patients whose radiological findings continued to deteriorate up to 24 months (Group B, n=21). The MRC scores of patients in Group A showed an early deterioration whereas in Group B, the MRC scores showed a gradual but unremitting deterioration. A similar evolution pattern was seen with FEV 1 /FVC% (Figure 2). These findings are indicative of an evolution of RILD from an early acute inflammation phase, characterised by reversible parenchymal change, into chronic inflammation, characterised by progressive lung volume loss and anatomical distortion.

Conclusion margin reduction does not underdoes clinical target volume in the majority of patients. GTV of primary tumour demonstrated a mean shrinkage of 48%. These two observation lead us to explore the implementation of a complete adaptive replanning strategy in the second or third week of treatment, with the aim of a better target coverage, sparing lung tissue and, if so, attempt to a isotoxic dose escalation. PD-0417 The evolution of radiation-induced lung damage following dose-escalated chemo-radiotherapy E. Chandy 1 , J. Jacob 2 , N. Yip 2 , A. Szmul 2 , D. Landau 1 , J.R. McClelland 2 , C. Veiga 2 1 University College London, UCL Cancer Institute, London, United Kingdom ; 2 University College London, Centre for Medical Image Computing, London, United Kingdom Purpose or Objective Radiation-induced lung damage (RILD) is a common consequence of lung cancer radiotherapy (RT) but its evolution over time is unclear. We investigate and quantify

Figure 1.

Figure 2.

Conclusion This study investigates the evolution of RILD using quantitative information on a set of homogeneously