Dosimetric evaluation of lung tumor immobilization using breath hold at deep inspiration

To examine the dosimetric benefit of self-gated radiotherapy at deep-inspiration breath hold (DIBH) in the treatment of patients with non-small-cell lung cancer (NSCLC). The relative contributions of tumor immobilization at breath hold (BH) and increased lung volume at deep inspiration (DI) in sparing high-dose lung irradiation (> or = 20 Gy) were examined.

Ten consecutive patients undergoing radiotherapy for Stage I-IIIB NSCLC who met the screening criteria were entered on this study. Patients were instructed to BH at DI without the use of external monitors or breath-holding devices (self-gating). Computed tomography (CT) scans of the thorax were performed during free breathing (FB) and DIBH. Fluoroscopy screened for reproducible tumor position throughout DIBH, and determined the maximum superior-inferior (SI) tumor motion during both FB and DIBH. Margins used to define the planning target volume (PTV) from the clinical target volume included 1 cm for setup error and organ motion, plus an additional SI margin for tumor motion, as determined from fluoroscopy. Three conformal treatment plans were then generated for each patient, one from the FB scan with FB PTV margins, a second from the DIBH scan with FB PTV margins, and a third from the DIBH scan with DIBH PTV margins. The percent of total lung volume receiving > or = 20 Gy (using a prescription dose of 70.9 Gy to isocenter) was determined for each plan.

Self-gating at DIBH was possible for 8 of the 10 patients; 2 patients were excluded, because they were not able to perform a reproducible DIBH. For these 8 patients, the median BH time was 23 (range, 19-52) s. The mean percent of total lung volume receiving > or = 20 Gy under FB conditions (FB scan with FB PTV margins) was 12.8%. With increased lung volume alone (DIBH scan with FB PTV margins), this was reduced to 11.0%, tending toward a significant decrease in lung irradiation over FB (p = 0.086). With both increased lung volume and tumor immobilization (DIBH scan with DIBH PTV margins), the mean percent lung volume receiving > or = 20 Gy was further reduced to 8.8%, a significant decrease in lung irradiation compared to FB (p = 0.011). Furthermore, at DIBH, the additional benefit provided by tumor immobilization (i.e., using DIBH instead of FB PTV margins) was also significant (p = 0.006). The relative contributions of tumor immobilization and increased lung volume toward reducing the percent total lung volume receiving > or = 20 Gy were patient specific; however, all 8 of the patients analyzed showed a dosimetric benefit with this DIBH technique.

Compared to FB conditions, at DIBH the mean reduction in percent lung volume receiving > or = 20 Gy was 14.3% with the increase in lung volume alone, 22.1% with tumor immobilization alone, and 32.5% with the combined effect. The dosimetric benefit seen at DIBH was patient specific, and due to both the increased lung volume seen at DI and the PTV margin reduction seen with tumor immobilization.