< Back to topics
> Cover Page, Authors, Institutions
> View tables/images/videos
Fourteen patients with stage IV breast or lung cancer (and with brain metastases) have been treated. The entire intracranial contents (to the inner table of the skull, and ~1 cm inferior the foramen magnum) were defined as the clinical target volume (CTV) in all cases (see Figure 1). The planning target volume (PTV) was defined as the CTV with a 2 mm uniform expansion margin. Every patient, for every treatment fraction, was positioned using image guided radiotherapy (IGRT) technique (stereoscopic kV X-ray/CT DRR fusion coupled with infrared fiducial marker matching via BrainLAB ExacTrac X-ray 6D, BrainLAB AG, Heimstetten, Germany). This IGRT technique is assumed to routinely achieve <2 mm setup error for intracranial cases. The lenses of the eyes were contoured in the treatment planning system (Eclipse v184.108.40.206, Varian Medical Systems, Palo Alto, CA), as was the "scalp" (defined as the skin with 3 mm of underlying tissue, contoured as a "cap" from the canthi to the vertex; see Figure 2). A 3-field beam arrangement was used: roughly opposed laterals (gantry at 85-95 or 265-275 degrees, with +/- 0-5 degrees of couch kick--achieving perfect lens superimposition for maximal lens sparing from each beam's eye view) as well as a "hemivertex" field (couch rotated at 90 degrees and gantry rotated 35-55 degrees to provide maximal lens-to-brain distance at a given beam's eye view). See Figure 3 for beam schema/beam's eye views/generated fluences.
Inverse optimization was used to generate IMRT plans: 100% of the dose was prescribed to 95% of the PTV and 99% of the CTV, and maximum total dose inhomogeneity was specified at 105%. Doses of 30 Gy/10 fractions (9 patients) or 37.5 Gy/15 fractions were used. A calculation grid size of 2.5 mm was the default for all plans, and inhomogeneity corrections (modified Batho) were used in all cases. For comparison, equal-dose standard plans using parallel opposed beam arrangements (open fields, rotated collimator, with skin flash) were generated for each patient (also calculated with 2.5 mm grid and inhomogeneity correction).