Intensity modulation to improve dose uniformity with tangential breast radiotherapy: Initial clinical experience

Purpose: We present a new technique to improve dose uniformity and potentia lly reduce acute toxicity with tangential whole-breast radiotherapy (RT) us ing intensity-modulated radiation therapy (IMRT). The technique of multiple static multileaf collimator (sMLC) segments was used to facilitate IMRT. Methods and Materials: Ten patients with early-stage breast cancer underwen t treatment planning for whole-breast RT using a new method of IMRT. The th ree-dimensional (3D) dose distribution was first calculated for equally wei ghted, open tangential fields (i.e., no blocks, no wedges). Dose calculatio n was corrected for density effects with the pencil-beam superposition algo rithm. Separate MLC segments were constructed to conform to the beam's-eye- view projections of the 3D isodose surfaces in 5% increments, ranging from the 120% to 100% isodose surface. Medial and lateral MLC segments that conf ormed to the lung tissue in the fields were added to reduce transmission. U sing the beam-weight optimization utility of the 3D treatment planning syst em, the sMLC segment weights were then determined to deliver the most unifo rm dose to 100 reference points that were uniformly distributed throughout the breast. The accuracy of the dose calculation and resultant IMRT deliver y was verified with film dosimetry performed on an anthropomorphic phantom. For each patient, the dosimetric uniformity within the breast tissue was e valuated for IMRT and two other treatment techniques. The first technique m odeled conventional practice where wedges were derived manually without con sideration of inhomogeneity effects (or density correction). A recalculatio n was performed with density correction to represent the actual dose delive red. In the second technique, the wedges were optimized using the same beam -weight optimization utility as the IMRT plan and included density correcti on. All dose calculations were based on the pencil-beam superposition algor ithm. Results: For the sMLC technique, treatment planning required approximately 60 min. Treatment delivery (including patient setup) required approximately 8-10 min. Film dosimetry measurements performed on an anthropomorphic phan tom generally agreed with calculations to within +/- 3%. Compared to the we dge techniques, IMRT with sMLC segments resulted in smaller "hot spots" and a lower maximum dose, while maintaining similar coverage of the treatment volume. A median of only 0.1% of the treatment volume received greater than or equal to 110% of the prescribed dose when using IMRT versus 10% with st andard wedges. A total of 6-8 segments were required with the majority of t he dose delivered via the open segments. The addition of the lung-block seg ments to IMRT was of significant benefit for patients with a greater propor tion of lung parenchyma within the irradiated volume. Since August 1999, 32 patients have been treated in the clinic with the IMRT technique. No patie nt experienced RTOG grade III or greater acute skin toxicity. Conclusion: The use of intensity modulation with an sMLC technique for tang ential breast RT is an efficient and effective method for achieving uniform dose throughout the breast. It is dosimetrically superior to the treatment techniques that employ only wedges. Preliminary findings reveal minimal or no acute skin reactions for patients with various breast sizes. (C) 2000 E lsevier Science Inc.