The use of simulator and CT information in the planning of radiotherapy for non-small cell lung cancer: An Australasian patterns of practice study

In a patterns of practice study 14 Australasian radiation oncologists were invited to define tumour and target volumes in 12 sample cases of non-small cell carcinoma of the lung (NSCCL) firstly using orthogonal simulator AP and lateral radiographic views, then using computed tomography (CT) slices obtained in the treatment position. In addition, they were asked to complete questionnaires addressing their criteria for determining treatment policy and their reasons for choosing specific tumour and target volume boundaries in individual cases. Significant variations in choice of size and position of both tumour and target volume were apparent between clinicians. These variations, however were no greater for simulator planned volumes than for CT planned volumes, except in cases selected for palliative treatment, were CT planned volumes were significantly larger. Failure to include tumour extensions, the mistaken inclusion of normal structures as tumour, and major differences in allowance for microscopic tumour spread were identified as the most significant cause for variation in the size and positioning of target volumes between clinicians. Differences over margins to allow for factors such as variation in day-to-day set-up, patient movement, equipment-related factors, etc, were also apparent. CT and simulator planned target volumes were equally successful in adequate tumour coverage; 79162 (49%), and 80162 (49%) respectively. Gross miss of tumour was more common in simulator planned target volumes than CT planned target volumes; 70162 (43%) and 55162 (34%), respectively. This study supports the integration of diagnostic expertise into the planning process for simulator and CT planning, but suggests that substantial practice variation would still exist in planning for radiotherapy in NSCCL.