PTV margin analysis for prostate patients treated with initial pelvic nodal IMRT and prostate proton boost

To evaluate the planning target volume (PTV) margins for prostate cancer patients treated with initial intensity-modulated radiation therapy (IMRT) to pelvic lymph nodes and a proton boost to the prostate using cone-beam computed tomography (CBCT) and radiographs as image guidance. We analyzed data from prostate patients (n  =  15) treated with initial pelvic IMRT and a proton boost to the prostate. CBCT and 2-dimensional (2D) radiographs were used for IMRT and proton pre-treatment alignments, respectively. Translations from bony-matching, implanted marker seed matching, and seed-bony differences were obtained from CBCTs. Pretreatment and post-treatment 2D radiographs were analyzed for residual setup errors and intrafractional organ motion. PTV margins of prostate and pelvic lymph nodes for seed-matching, bony-matching and the skin mark-matching approach were obtained using 2.5 * Σ  +  0.7 * σ formula, where Σ is the systematic error and σ is the random error summed in quadrature of various error components. For the seed-matching approach, the prostate margins were 1.5 mm, 3.5 mm, and 2.7 mm for the left-right (LR), superior-inferior (SI), and anterior-posterior (AP) directions, respectively; pelvic lymph node margins were 2.3 mm, 7.1 mm, and 7.0 mm. For the bony-matching approach, the prostate margins were 2.7 mm, 7.8 mm, and 7.5 mm for the LR, SI, and AP directions, respectively; pelvic lymph node margins were 0.7 mm, 1.5 mm, and 1.4 mm. For the skin mark-matching approach, the prostate margins were 7.6 mm, 13.4 mm, and 11.4 mm for the LR, SI, and AP directions, respectively; pelvic lymph node margins were 8.2 mm, 12.6 mm, and 12.3 mm. The seed-matching approach, as compared to the bony-matching approach, reduced prostate margins, which may in turn reduce dose to the bladder and rectum at the expense of increased but acceptable margins for the pelvic lymph nodes.