A technique of partial breast irradiation utilizing proton beam radiotherapy: Comparison with conformal X-ray therapy

Purpose: To develop a breast immobilization system and clinical technique to deliver partial breast irradiation with a proton beam and compare dose distributions using proton therapy and conformal x-rays. Methods: A clinical technique to provide reproducible breast immobilization was developed. Breast immobilization begins by fitting each patient with a treatment brassiere. Patients are placed prone in a cylindrical polyvinyl chloride shell with the upper and lower body being supported and immobilized with Vac-Lok foam bead cushions. The upper chest and breast areas are immobilized with two-part expandable foam. After a treatment planning computed tomography scan, the lumpectomy cavity is outlined, and a clinical target volume is generated by adding I cm in all dimensions. A three-dimensional treatment plan is developed with treatment typically given with 2 to 4 separate proton beams. The dose administered is 40 cobalt Gray equivalents (CGE) delivered in 10 daily fractions of 4 CGE with multiple fields treated each day. Proton and conformal x-ray plans were compared using dose-volume histogram analysis to determine volumes of normal breast tissue and skin treated with each technique. Results: An institutional review board-approved clinical trial was developed using this technique, and 20 patients have completed treatment. All subjects were able to undergo the immobilization procedure and daily treatments without significant discomfort, and no treatment interruptions were encountered. There was no evidence of respiratory motion identified on treatment planning CT images or on daily set-up radiographs. Acute toxicity has been limited to occasional radiation dermatitis (Radiation Therapy Oncology Group grade 1-2). Proton plans were compared with 2 methods of photon partial breast irradiation, including reduced tangential fields and five-field conformal techniques. Ten treatment plans with dose-volume histogram analysis revealed that the use of proton beams provided a significant reduction in doses to the ipsilateral breast and skin while eliminating doses to the heart and lung tissues. Conclusions: A simple immobilization procedure provides accurate and reproducible breast positioning while simultaneously eliminating respiratory motion. The procedure has been well tolerated by the first 20 patients. Protons can provide substantial normal tissue protection compared with the use of conformal x-rays when used for partial breast treatment. We plan to continue enrollment and analyze long-term toxicity, local control, and survival.