ABSTRACT
AIMS: To study the accuracy of tumour-volume localisation in a comparison of conventional and virtual simulation for palliative lung radiotherapy. To assess if three-dimensional tumour outlining is necessary for the virtual simulation process. MATERIALS AND METHODS: Ten consecutive patients with non-small cell lung cancer underwent target localisation for palliative lung radiotherapy using conventional and virtual simulation. The treatment fields were initially marked with a conventional simulator using fluoroscopy, plain X-ray film and available diagnostic imaging. Each patient then had a computed tomography (CT), and these simulated treatment fields were reproduced within the virtual simulation planning system. Two clinicians then independently defined treatment fields using virtual simulation alone. The virtual simulation was achieved without outlining the tumour in three dimensions. The coverage of an 'ideal' CT-defined planning-target volume (PTV) was then calculated for each of the virtually and conventionally simulated fields. In addition, the amount of irradiated normal lung was measured using dose-volume histograms (DVH). Field sizes and differences in tumour volume coverage were compared. RESULTS: There was significantly greater tumour volume coverage using virtual simulation compared with conventional simulation (P < 0.03). This advantage was more pronounced in tumours that were larger and those that were closer to the patient's midline. There was no statistically significant difference in the volume of uninvolved lung irradiated between the two methods. CONCLUSION: In this small sample of patients, we have demonstrated improved tumour volume coverage using virtual simulation, without increasing the volume of uninvolved lung treated. A simple but consistent method of virtual simulation for this patient group is offered as an alternative to both PTV-defined CT simulation and fluoroscopy-based conventional simulation.