ESTRO 36 Abstract Book

S392 ESTRO 36 _______________________________________________________________________________________________

Results From 2000 to 2011, 121 patients (median age: 50 years, range 16-86; median follow-up: 54 months), affected by primary high grade STS, underwent unplanned surgery, re- excision of the tumor bed (radicalization) within a maximum of 3-6 months from the previous surgery, perioperative BRT and adjuvant EBRT. Seventeen patients (14.0%) developed metastases, 7 patients (5.8%) relapsed and 9 out of 121 patients died (7.4%). Five-year LC and OS were 93.0% and 91.6%, respectively. At univariate analysis higher 5-year DFS and OS were recorded in patients with lower- limb tumors vs upper-limb and trunk STS (p: 0.053 and 0.041, respectively). Although it wasn’t detected any statistical significance related to histologies. Younger patients (< median age) showed improved 5-year LC (97.9% vs 88.1%, p: 0.052), 5-year DFS (88.9% vs 73.9%, p: 0.034) and 5-year OS (96.5% vs 86.6%, p: 0.093). Conclusion The combination of BRT and EBRT is able to achieve satisfactory results, with a high local control rate and overall survival. Prospective studies on combined modality treatment in the adjuvant setting of STS after re-excision surgery or inadequate excision are still needed to improve the results in STS of the trunk and limb. PO-0745 Intrafractional movement of patients with spinal cord compression receiving radiation therapy E. Ravnkilde 1 , I.M. Jakobsen 1 , A. Appelt 1 , J.P. Bangsgaard 1 , L.S. Fog 1 1 Rigshospitalet, Clinic of Oncology, Copenhagen, Denmark Purpose or Objective Many palliative radiation therapy patients experience moderate to severe pain. This pain could potentially increase intrafractional movement, requiring planning treatment volume (PTV) margins to be adjusted to account for this. We conducted a prospective study to examine the impact of patient-experienced pain on intrafractional movement and the time needed for treatment delivery. Material and Methods This prospective study included 18 consecutive patients receiving radiation therapy for spinal cord compression. We recorded the patients’ intrafractional shifts, the treatment time, the treatment site and the patients’ self- reported pain score. The patients were asked to assess their pain (‘pain’) prior to radiation therapy fraction, on a scale from 1 to 10 using the NRS (Numerical Ratings Scale) [Pain Pract. 3 (4): 310–6] . Cone beam CT images were acquired before and after all daily treatments. The interfractional shift (‘shift’), linac ID number (‘linac’), treatment time (‘time’), fraction number (‘’fr number’) and treatment site (‘site’) were recorded. The average and maximum shifts, and the standard deviation (s.d.), were determined. Spearman correlation coefficients were calculated between: shift and time, fr number, or pain; time and fr number or pain; pain and fr number. Since site was scored by a categorical variable, a Kruskal Wallis test was used to investigate effect of treatment site on shift, time or pain score. The patients provided informed consent to participation in the study. Results A total of 113 shifts were measured. The average shift was 0.96 mm, the maximum 4.1mm, and the s.d. was 0.89 mm (fig 1). The only significant correlations (see fig 2) were between - Shift and pain (patients reporting more pain had greater shifts, p=0.0045, r=0.2699). Patients who report more pain Poster: Clinical track: Palliation

have greater intrafractional shifts. This may be due to patient discomfort. - Time and fr number (later fractions were completed more quickly, p=0.0001, r=0-0.3500). The reduced time for later fractions may be due to the patient becoming more familiar with the treatment procedure. - Pain and fr number (patients reported less pain in later fractions , p=0.0412, r=-0.1960). While pain score decreased with fraction number, fewer patients provided pain scores for later fractions. - Site and time (p=0.0044, C²=10.87). Treatment site correlates with treatment time. Patients with pain scores £5 had mean intrafractional shift 0.09 cm (s.d. 0.09), while patients with pain scores >5 had mean shift 0.11 (s.d.0.07).

Conclusion A 5 mm PTV margin appears sufficient to account for intrafractional patient movement for spinal cord compression patients imaged daily. PO-0746 Inter-observer variation in GTV delineation of bone metastases: a multicenter study A.S. Gerlich 1 , J.M. Van der Velden 1 , A.N.T.J. Kotte 1 , C.L. Tseng 2 , G. Fanetti 3 , W.S.C. Eppinga 1 , N. Kasperts 1 , M.P.W. Intven 1 , F.A. Pameijer 4 , M.E.P. Philippens 1 , H.M. Verkooijen 5 , E. Seravalli 1 1 University Medical Center Utrecht, Radiation Oncology, Utrecht, The Netherlands 2 Sunnybrook Health Sciences Centre, Radiation Oncology, Toronto, Canada 3 European Institute of Oncology, Radiation Oncology, Milan, Italy 4 University Medical Center Utrecht, Radiology, Utrecht, The Netherlands 5 University Medical Center Utrecht, Imaging, Utrecht, The Netherlands