No abstract

Objectives: Implement scripted automatic breast planning (AP) for breast techniques within Raystation. Methods: Manual plans (MP) were re-planned and compared with AP plans for whole breast (WB), partial breast (PB), hybrid VMAT SIB and VMAT nodal plans. Results: WB AP plans took 7 min comparing well to MP. One WB AP failed a mandatory dose constraint. Small statistically significant differences showed improved coverage for AP at expense of slightly hotter plans, however absolute differences were small (mean differences < 1% or D 0.5cc <0.2 Gy). PB AP plans took 9 min, showing improved coverage (V 24.7Gy97.6versus 96.4 %). One PB AP case failed a mandatory constraint. Other dosimetric differences were non-significant. SIB AP plans took 14 mins with one case failing a mandatory constraint with minor differences compared with MP except larger V 42.8Gy (3vs 1.5 %) and more μ. VMAT AP plans took 12 mins and were hotter for PTVp_4000 but had higher nodal coverage. Contra_Lung V 2.5Gy was higher (8.8 %) than MP plans (6.5 %). Conclusion: Automatic planning of modern breast techniques has been successfully introduced using a commercial planning system. AP plans are very similar to MP, requiring little manual interaction for most cases with significant timesaving potential. Advances in knowledge: Scripted breast plans produced within minutes for WB, PB, SIB and VMAT. Successfully introduced into large busy department. Plans similar to manual plans, requiring little manual interaction.

No abstract

AimsCraniospinal irradiation (CSI) remains a crucial treatment for patients with medulloblastoma. There is uncertainty about how to manage meningeal surfaces and cerebrospinal fluid (CSF) that follows cranial nerves exiting skull base foramina. The purpose of this study was to assess plan quality and dose coverage of posterior cranial fossa foramina with both photon and proton therapy.Materials and methodsWe analysed the radiotherapy plans of seven patients treated with CSI for medulloblastoma and primitive neuro-ectodermal tumours and three with ependymoma (total n = 10). Four had been treated with a field-based technique and six with TomoTherapy™. The internal acoustic meatus (IAM), jugular foramen (JF) and hypoglossal canal (HC) were contoured and added to the original treatment clinical target volume (Plan_CTV) to create a Test_CTV. This was grown to a test planning target volume (Test_PTV) for comparison with a Plan_PTV. Using Plan_CTV and Plan_PTV, proton plans were generated for all 10 cases. The following dosimetry data were recorded: conformity (dice similarity coefficient) and homogeneity index (D2 − D98/D50) as well as median and maximum dose (D2%) to Plan_PTV, V95% and minimum dose (D99.9%) to Plan_CTV and Test_CTV and Plan_PTV and Test_PTV, V95% and minimum dose (D98%) to foramina PTVs.ResultsProton and TomoTherapy™ plans were more conformal (0.87, 0.86) and homogeneous (0.07, 0.04) than field-photon plans (0.79, 0.17). However, field-photon plans covered the IAM, JF and HC PTVs better than proton plans (P = 0.002, 0.004, 0.003, respectively). TomoTherapy™ plans covered the IAM and JF better than proton plans (P = 0.000, 0.002, respectively) but the result for the HC was not significant. Adding foramen CTVs/PTVs made no difference for field plans. The mean Dmin dropped 3.4% from Plan_PTV to Test_PTV for TomoTherapy™ (not significant) and 14.8% for protons (P = 0.001).ConclusionsHighly conformal CSI techniques may underdose meninges and CSF in the dural reflections of posterior fossa cranial nerves unless these structures are specifically included in the CTV.