Purpose: The purpose of this work was to assess the magnitude of setup uncertainties and respiratory-induced motion of lung tumors by monitoring the location of fiducials implanted in the vicinity of the tumors.
Methods and materials: Gold fiducials were implanted in the periphery of lung tumors in 5 patients who had Stage III non-small-cell lung cancer. Fiducial motion was measured using weekly repeated four-dimensional computed tomography (4DCT) imaging and during gated treatment each day using an electronic portal imaging device (EPID). Setup uncertainties were quantified using both the EPID images and the 4DCT data sets.
Results: We observed a reduction in fiducial motion (left/right and superior/inferior directions) during gated treatment; however, large gated motion was present (>1 cm). Systematic and random uncertainties based on patient setup ranged from 4 to 6 mm in all three directions as measured using fiducials on gated EPID images and repeat 4DCTs, and using bony anatomy on repeat 4DCTs.
Conclusions: Respiratory gating may be an effective method of reducing average motion during the course of treatment, but large motion is still possible when delivering gated treatment. Setup uncertainties were on the order of, if not larger than, residual gated motion. We recommend careful consideration of all sources of error before reducing margins on the basis of respiratory motion management alone without a strategy for accurate patient setup on a daily basis.