Abstract Book

S1102

ESTRO 37

Purpose or Objective Thoracic and abdominal treatment sites can be affected by respiratory motion. When targeting with radiotherapy, it is important that this respiratory motion is accounted for. This is typically done by creating an ITV, determined by a 4DCT or extreme-phase breath-hold CT (i.e. end- expiration (EEBH) and inspiration (IBH) CT). However, this approach can often lead to large volumes as the ITV needs to cover the entire extent of respiratory motion. Alternative approaches to minimise the respiratory motion include a breath-hold technique. The breath-hold from CT is replicated throughout treatment, and the uncertainty in breath-hold position is incorporated into a CTV–PTV margin. This study investigates a motion management strategy which aims to provide the optimal motion management technique for each individual patient. Figuratively, this strategy is shown in figure 1.

significant volume reduction was achieved by using the alternative technique. Therefore it's important to have a flexible approach to motion management. EP-2020 Deep inspiration breath hold and locally advanced lung cancer: validation of geometrical uncertainties M. Josipovic 1 , M. Aznar 2 , S. Damkjær 1 , J. Thomsen 1 , J. Rydhög 3 , L. Nygård 1 , L. Specht 1 , M. Pøhl 1 , G. Persson 1 1 The Finsen Center - Rigshospitalet, Dept. of Oncology- Section of Radiotherapy, Copenhagen, Denmark 2 University of Manchester, Division of Cancer Sciences- School of Medical Sciences - Faculty of Biology- Medicine and Health, Manchester, United Kingdom 3 Skånes University Hospital Lund, Department of Radiation Physics, Lund, Sweden Purpose or Objective Uncertainties in deep inspiration breath hold (DIBH) radiotherapy (RT) for locally advanced non-small cell lung cancer (NSCLC) have previously been evaluated in a pilot study of 17 patients, extensively imaged in DIBH, but treated in free breathing [Ref XXX]. Based on these uncertainties adequate planning target volume (PTV) margins were established prior to initiation of a larger prospective trial, treating patients in DIBH. The purpose of the current work is to validate the findings of the previously published pilot study using patient data from our prospective DIBH trial. Material and Methods Patients were included from 2015-2017 in a prospective DIBH trial at a single institution. Respiratory coaching with visual guidance was used to ensure comfortable voluntary DIBHs for RT imaging and delivery. Patients underwent 3 consecutive CT scans in 3 consecutive visually guided DIBHs as part of imaging for RT planning. Second and third DIBH CTs were rigidly registered onto the first DIBH CT, based either on primary tumour (T), mediastinal lymph nodes (N) or carina (C). The following inter-DIBH uncertainties were evaluated: position of the T, N and C, differential motion between T & N and uncertainty of using C as a surrogate for N during daily CBCT image guidance. PTV margins were calculated based on [van Herk 2004]. Clinical parameters, DIBH related uncertainties and PTV margins from DIBH trial were compared to the pilot study. Non-parametric statistics were applied. Results DIBH CT scan sets from 61 included patients were available. Two patients were not deemed compliant for RT in DIBH (tumour position deviations between consecutive DIBHs >1cm) and were excluded from the analysis. Final analysis was performed on 59 patients. Median lung volume increase in DIBH was 54% in the DIBH trial, which did not differ significantly from the 60% increase in the pilot study. Median age in both studies was 67 years. GTV size, gender, performance status, smoking status and clinical disease stage did not differ significantly between the studies (see Table 1). The only clinical parameter differing significantly between the studies was forced expired volume in 1 second (FEV1%), which was higher in the DIBH trial group (81% vs 64%, p=0.04). 3D inter-DIBH uncertainties in DIBH trail were small: 1.9±1.6 mm (mean ± SD) for T and 1.7±1.4mm for N. Intra-fractional 3D differential motion between T and N was 0.4±1.6 mm. Table 2 presents details for cardinal axes and comparison to the pilot study data. Geometrical uncertainties, evaluated in both studies, did not differ significantly; however the resulting PTV margins differed (Table 2). The PTV based on DIBH trial was reduced by ~1mm in two directions and increased by1.5mm in left- right for N and by ~1mm in one or two directions for T.

Material and Methods 43 patients who were assessed within our motion management strategy were reviewed. The patients included several abdominal and thoracic sites (predominantly pancreas, liver, lung, oesophagus, and mediastinum). For each patient, PTVs were generated using both ITV (PTV ITV ) and breath-hold (PTV BH ) techniques. For ITV, a 5mm PTV margin was applied. For breath-hold, the breath-hold uncertainty during treatment was assessed and included on top of 5mm in the CTV-PTV margin. The volumes of the PTVs generated by both motion management techniques were compared. Results The difference in the PTV volumes between the two techniques varied, with a mean volume difference for all patients of 51cc (15% relative change) with a standard deviation of 52cc. For pancreas, the PTV BH was smaller in 7/11 patients with a mean reduction of 60cc (28.4%) and maximum of 143cc (35%). For 4/11 patients, the PTV ITV was smaller by a mean of 23cc (17%) and a maximum of 64cc (13%). For liver, the PTV BH was smaller in 6/7 patients with a mean reduction of 93cc (14.9%) and maximum of 189cc (38.7%). For 1/7 patients, the PTV ITV was smaller by a 89cc (23.5%). For lung, the PTV BH was smaller in 4/7 patients with a mean reduction of 15cc (13.9%) and maximum of 38cc (46%). For 3/7 patients, the PTV ITV was smaller by a mean of 27cc (10.8%) and a maximum of 51cc (18.8%). For oesophagus, the PTV BH was smaller in 2/6 patients with a mean reduction of 120cc (19.7%) and maximum of 124cc (23.9%). For 4/6 patients, the PTV ITV was smaller by a mean of 52cc (14.6%) and a maximum of 85cc (10.4%). For mediastinum, the PTV BH was smaller in 1/5 patients with a reduction of 180cc (15.3. For 4/5 patients, the PTV ITV was smaller by a mean of 12cc (4.5%) and a maximum of 14cc (5.6%). Conclusion The results show that the optimal motion management strategy to minimise the irradiated volume is patient- specific. While liver (PTV BH ) and mediastinum (PTV ITV ) both showed a clear trend, both sites had a case where