Abstract Book

S1076

ESTRO 37

EP-1978 Lung SABR: New optimisation techniques S. Currie 1 , G. Currie 1 , P. Houston 1 1 NHS greater glasgow and clyde, RT Physics, glasgow, United Kingdom Purpose or Objective To investigate whether utilising Eclipse TPS v15.5 [Varian Medical Systems, Palo Alto, CA, USA] multi-criteria optimisation (MCO) can produce improved treatment plans, in terms of reduction in dose to organs at risk and/or improved planning target volume coverage, for VMAT radiotherapy treatment of Lung SABR when compared with a plan generated using RapidPlan TM alone. Material and Methods A retrospective planning study was performed for ten lung SABR patients. Each patient treatment was planned as standard using RapidPlan TM . The plans were then re- optimised within the Eclipse Treatment Planning v15.5 MCO tool (E-MCO). Both plans were normalised to ensure that 95% of the target received 100% of the prescribed dose and the resulting plans were then compared. Differences in dose volume histogram parameters were calculated to assess plan quality. Significance was assessed by two-tailed t-test (p<0.01). Results The E-MCO generated plans exhibited significant organ at risk (OAR) reductions for the dose at 2cm from PTV (mean: 12.9%; range: 9.6% - 24.0%), the maximum spinal cord dose (mean: 3.2 Gy; range: 0.3 Gy – 6.7 Gy), the dose to 1cc of oesophagus (mean: 1.2 Gy; range: 0.3 Gy – 3.2 Gy) and percentage of lung minus GTV to receive 20Gy (mean: 12.8%; range: 0.0% - 25.0%). There was no significant difference observed for the remaining organs at risk. As the plans were normalised to achieve the same target coverage, there was no significant difference in PTV coverage. The plans generated by both RapidPlan TM and RapidPlan TM with E-MCO were all clinically acceptable and met local OAR and PTV dosimetric objectives. Conclusion Using E-MCO to re-optimise RapidPlan TM generated plans was shown to decrease significantly the dose to organs at risk without adverse affect on tumour coverage. E-MCO demonstrated the ability to further enhance plans that were already of a high standard. EP-1979 Automated non-inferiority validation of knowledge-based planning across multiple disease sites R. Kaderka 1 , R. Mundt 1 , N. Li 1 , B. Ziemer 1 , T. Atwood 1 , M. Cornell 1 , K. Moore 1 1 University of California San Diego, Radiation Medicine and Applied Sciences, San Diego, USA Purpose or Objective Knowledge-based planning (KBP) studies seek to demonstrate potential quality and efficiency gains, yet broad clinical implementation of KBP necessitates significant up-front effort even within a single disease. In this study, patient plans treated just before multi-site implementation of KBP were batch re-planned using an automated workflow to (a) assess the non-inferiority of KBP and (b) estimate potential clinical gains post- implementation. Material and Methods We identified 51 prostate, 24 prostatic fossa, 54 hypo- fractionated lung (27 left, 27 right) and 30 head-and-neck patients treated in the year directly preceding broad clinical adoption of KBP at our clinic. Within these disease sites, we presently have clinical KBP models developed in Varian RapidPlan (Varian Medical Systems, Palo Alto, CA, USA). Using the v15.1 Eclipse Scripting API (ESAPI), we developed a program that automatically batch processes multiple patients and generates KBP plans, taking as input a list of patients to be re-planned, then performs semi-automated structure matching and

respectively (.006). This superior coverage could be achieved without significantly increasing dose to the bladder, rectum and sigmoid.

Conclusion GEC ESTRO based contouring guidelines cover all the functionally abnormal areas on DWI. In this study DWI based plans when compared to T2W plans resulted in smaller HRCTV and IRCTV volumes leading to better coverage without increasing dose to the organs at risk. But care must be taken in using DWI as a standalone imaging for planning and contouring, due to its poor anatomic visualization and distortions there is a huge risk of geometric and applicator reconstruction uncertainties. If at all DWI has to be used for planning, image fusion with T2W is a must. Hence we recommend that DWI should only be used as a supplement to T2W for contouring target volumes and the plan should always be done on the T2W imaging as recommended by GEC ESTRO