Abstract book - ESTRO meets Asia

S82 ESTRO meets Asia 2018

1 Sir Run Run Shaw Hospital- School of Medicine- Zhejiang University-, Radiation Oncology, Hangzhou, China 2 Hangzhou Tumor Hospital, Radiotherapy Medical Physics Research Lab, Hangzhou, China 3 The First People’s Hospital of Hangzhou, Radioaation Oncology, Hangzhou, China 4 University of Nevada, Department of Medical Physics, Las Vegas, USA Purpose or Objective In this paper, we study the reliability of a novel deep transfer learning framework for internal gross target volume (IGTV) delineation from four-dimensional computed tomography (4DCT), which is applied to patients with lung cancer treated by Stereotactic Body Radiation Therapy (SBRT). Material and Methods 77 patients who underwent SBRT followed by 4DCT scans were incorporated in a retrospective study. The IGTV_DL was delineated using a novel deep machine learning algorithm with a linear exhaustive optimal combination framework, for the purpose of comparison, three other IGTVs base on common methods was also delineated, we compared the relative volume difference (RVI), matching index (MI) and encompassment index (EI) for the above IGTVs. Then, multiple parameter regression analysis assesses the tumor volume and motion range as clinical influencing factors in the MI variation. Results Experimental results demonstrated that the deep learning algorithm with linear exhaustive optimal combination framework has a higher probability of achieving optimal MI compared with other currently widely used methods.

30# starting day 1 @ 2Gy/# & 5#/week using CO-60 machine. In both the arms spinal cord off was done after 44Gy. Quality of life was evaluated and recorded weekly using EORTC QLQ–LC13 questionnaire. Improvement / deterioration in any scale was analysed between the baseline score and the score at treatment completion or 1 st follow-up, whichever was higher. Results The median age of patients were 56 years. Out of a total of 36 patients, 19 patients had stage III A disease and 17 patients had stage III B disease. 26 patients had squamous histology, 10 had adenocarcinoma histology. When the response rates at 1 st follow-up were assessed, CR was obtained in 2 patients in the control and 2 patients in the study arm. PR was obtained in 11 patients in the control arm (61.1%) and 13 patients in the study arm (72.2%). There were 2 patients in the control arm (11.1%) and 1 in the study arm(5.6%) who were found to have disease progression at 1 st follow up. The observation were statistically insignificant. When grade ˃ III toxicities are analysed, the total number of events of grade ˃ III toxicities in the control arm were 3(16.7%) and in the study arm were 5(27.8%), which is statistically insignificant. Conclusion The response rates and disease progression are similar between study and the control arm. The other two end points namely the toxicity profile and quality of life are also similar between the two arms. However larger study with longer follow up are needed to establish the comparability of these two regimens. It is recommended that the treatment options should be individualized. PO-202 Concomitant chemoradiotherapy vs accelerated rt in locally advanced non-small cell lung cancers S. rana 1 1 pgimer chandigarh, radiation oncology, new delhi, India Purpose or Objective To compare the disease response, loco-regional tumor control, quality of life and toxicity profile in accelerated radiation (six fractions per week) and conventional chemoradiation in locally advanced non-small cell lung cancer. Material and Methods Total 50 patients were enrolled and randomized into two groups the study (Accelerated Radiation n=25) and control group(Conventional chemoradiation with etoposide – cisplatin n=25 ). We included previously untreated patients of locally advanced inoperable or non resectable NSCLC. Patients assigned to accelerated radiation arm were given radiation six fractions per week (60Gy/5wks/30#) from Monday to Saturday. Patients assigned to Concomitant chemoradiation arm were given radiation 5 fractions per week (60Gy/6wks/30#) from Monday to Friday along with Injection cisplatin 20 mg/m 2 iv days 1-5 & days 29-33 + Injection etoposide 50 mg/m 2 iv days 1-5 & days 29-33. Results The response in both the treatment arms at 6 weeks follow up was comparable (p=0.569). The grade II hematological toxicity was more in control arm (p=0.000).The quality of life parameter like hair loss and sore throat worsened in control arm. Conclusion Accelerated RT may prove a good alternate to concurrent CRT in lung cancer patients who are not a suitable candidate for CRT. PO-203 A Novel Transfer Learning Framework for IGTV Definition from 4DCT of Lung Cancer Patients X. Li 1,2,3 , Q. Deng 3 , L. Zhang 2 , T. Niu 1 , Y. Kuang 4 , S. Wu 2 , S. Ma 3

MI distribution of IGTV10_DL in three subgroups with different tumor 3D motion range. (a) MI_IGTV10_DL R3D<5mm. (d) MI_IGTV10_DL R3D<5mm with optimal MI displayed. (b)MI_IGTV10_DL R3D: 5~10mm. (e) MI_IGTV10_DL R3D:5~10mm with optimal MI displayed. (c)MI_IGTV10_DL R3D>10mm. (f) MI_IGTV10_DL R3D>10mm with optimal MI displayed.

MI distribution of IGTV10_DEF in three subgroups with different tumor 3D motion range. (a) MI_IGTV10_DEF