ESTRO 38 Abstract book

S320 ESTRO 38

The agreement between the dose distributions in EBT3 film and calculated by the TPS was evaluated with FilmQA Pro software using triple channel dosimetry.

the results were subdivided by IMRT QA device, the recalculation had a significantly higher sensitivity than any specific device examined (Table). Specificity was lower for the independent recalculation, but overall, incorporating both sensitivity and specificity, ROC analysis found superior performance (statistically significant) for the independent recalculation over clinical measurement- based IMRT QA. The area under the curve was 0.79 for the independent recalculation but only 0.60 for the measurement based approaches. Based on the ROC curves, the independent recalculation could detect 80% of unacceptable plans using a dose disagreement criteria (between the recalculation and the TPS value) of 3.8%. In order for measurement based IMRT QA methods to detect 80% of unacceptable plans, a clinically unrealistic criteria of 99.7% of pixels would need to pass a 3%/3mm criterion.

Results Analysis of small field output factors calculated in the pre- visit phase showed that differences between participants’ data and the Imaging and Radiation Oncology Core, IROC – Houston QA Centre reference dataset were generally within the tolerance of 3% for all field sizes. Good agreement was found between calculated and film measured profiles for the 2x2 cm 2 field in all centres. MLC QA test results were within 0.5 mm for the leaf positioning bias in all participating institutions. All treatment plans have met the proposed plan constrains except three plans where at least one of the constraints was not accomplished. The differences between ionization chamber measurements and calculated doses were on average -0.1 ± 2.0 % in PTVs and 0.2 ± 2.2% in the spinal cord, with the individual results of all centres within the established tolerances of ± 5% for PTVs and ± 7% for the spinal cord. One follow up visit was required to resolve a major deviation in the spinal cord. Considering the results of film analysis, for global gamma criteria of 3%/3mm 20% TH, passing rates ranged from 90.3% to 99.1%, again all above the limit of acceptability of 90%. Conclusion The IMRT audit supported by the IAEA carried out in Portugal showed an overall good agreement between planned and delivered doses although the IMRT treatment planning exercise was considered demanding. OC-0608 Credentialing of spine stereotactic ablative body radiotherapy in a multi-centre trial N. Hardcastle 1 , O. Cook 2 , P. Mitchell 3 , S. Siva 4 1 Peter MacCallum Cancer Centre, Physical Sciences, Melbourne, Australia ; 2 Trans-Tasman Radiation Oncology Group, Quality Assurance, Newcastle, Australia ; 3 Olivia Newton John Cancer and Wellness Centre, Medical Oncology, Melbourne, Australia ; 4 Peter MacCallum Cancer Centre, Radiation Oncology, Melbourne, Australia Purpose or Objective The NIVORAD trial (ALTG14/002, ACTRN12616000352404) involves administration of immunotherapy with or without Stereotactic Ablative Body Radiotherapy (SABR) to a single extracranial lesion in metastatic non-small cell lung cancer (NSCLC). Treated lesions receive a single fraction of 18-20 Gy and can include lung, soft tissue and bone including spine. Given the multi-disciplinary and complex nature of SABR delivery in particular with spine, a credentialing approach was created focusing on review of SABR treatment processes and measurement of a spine case. Material and Methods Credentialing was split into two streams. Centres with a previous SABR site visit through TROG were required to perform an axial film measurement of a benchmarking spine plan in an appropriate local phantom and submit raw radiochromic film images for analysis. Remaining centres had an on-site review of SABR processes based on international guidelines, followed by an axial film measurement of the spine benchmarking plan. Films were

Recalcula tion specificit y (%)

Device

Measurement specificity (%)

Measurement sensitivity (%)

Recalculation sensitivity (%)

# tests #failing results

All

337

18 6

72 71

98

68 57 70

EPID

58

7 4

0 0

100 100

ArcCheck 93 Ion Chamber 44 MapCheck 121

100

1

0

100

91

67

5

20

40

100

67

Conclusion A simple independent recalculation using linac class- specific beam models had much better sensitivity at detecting unacceptable plans as compared to existing clinical measurement-based IMRT QA methods. OC-0607 IAEA supported national ‘end-to-end’ IMRT audit in Portugal T. Santos 1 , M.D.C. Lopes 2 , E. Gershkevitsh 3 , J. Izewska 4 1 Faculty of Sciences and Technology, Physics Department, Coimbra, Portugal ; 2 Instituto Portugues de Oncologia de Coimbra Francisco Gentil E.P.E., Medical Physics Department, Coimbra, Portugal ; 3 North Estonia Medical Centre, Medical Physics Department, Tallinn, Estonia; 4 International Atomic Energy Agency, Dosimetry and Medical Radiation Physics Section, Vienna, Austria Purpose or Objective IAEA has developed an IMRT audit programme to review physical aspects of IMRT H&N treatments through on-site visits. The audit was carried out in Portugal between April and September 2018. This abstract presents the results. Material and Methods All radiotherapy centres performing IMRT treatments in Portugal, 20 out of 24 centres, have voluntarily participated. The centres equipment to perform IMRT is shown in Table 1. Using a specially designed anthropomorphic H&N phantom – SHANE (CIRS Norfolk, VA) – and a set of contours, the audit methodology simulates all steps of a nasopharynx IMRT treatment, from CT scanning to treatment delivery, following the local protocol. To guide the planning optimization, a list of dose objectives and constraints was provided. As part of the auditing process, pre-visit activities were performed concerning small field dosimetry and MLC performance which were also checked during the on-site visit. Beam output fluctuation on the audit day was taken into account. The dosimetric verification of the H&N IMRT plan was performed with a dedicated dosimetry set including a small volume ionization chamber placed inside the phantom at four different locations (three PTVs and spinal cord) and an EBT3 film to assess the dose distribution in a coronal plane.