Deep inspiration breath-hold technique for lung tumors: the potential value of target immobilization and reduced lung density in dose escalation

Purpose/Objective: Lung tumors are subject to movement due to respiratory motion. Conventionally, a margin is applied to the clinical target volume (CTV) to account for this and other treatment uncertainties. The purpose of this study is to evaluate the dosimetric benefits of a deep inspiration breath-hold (DIBH) technique which has two distinct features - deep inspiration which reduces lung density and breath-hold which immobilizes lung tumors. Both properties can potentially reduce the mass of normal lung tissue in the high dose region, thus improving the possibility of dose escalation. Methods and Materials: To study the efficacy of the DIBH technique, CT scans are acquired for each patient under 4 respiration conditions: free-breathing; DIBH; shallow inspiration breath-hold; shallow expiration breath-hold. The free-breathing and DIBH scans are used to generate treatment plans for comparison of standard and DIBH techniques, while the shallow inspiration and expiration scans provide information on the maximum extent of tumor motion under free-breathing conditions. To acquire the breath-hold scans, the patients are brought to reproducible respiration levels using spirometry and slow vital capacity maneuvers. For the treatment plan comparison free-breathing and DIBH planning target volumes (PTVs) are constructed consisting of the CTV plus a margin for setup error and lung tumor motion. For both plans the margin for setup error is the same while the margin for lung tumor motion differs. The margin for organ motion in free-breathing is determined by the maximum tumor excursions in the shallow inspiration and expiration CT scans. For the DIBH, tumor motion is reduced to the extent to which DIBH can be maintained and the margin for any residual tumor motion is determined from repeat fluoroscopic movies, acquired with the patient monitored using spirometry. Three-dimensional treatment plans, generated using apertures based on the free-breathing and DIBH PTVs, are compared to assess the improvement due to target immobilization and reduced lung density. To estimate the role of target immobilization alone, an additional plan is generated using the DIBH PTV, but with the assumption it is surrounded by free-breathing, not deep inspiration, density lung. Results: The results of the study suggest that the DIBH technique can reduce the mass of lung irradiated to high dose, so that the possibility for dose escalation is increased. The relative contribution of reduced lung density and reduced margin for motion vary depending on the tumor size, level of DIBH and extent of tumor motion in free-breathing. An example treatment plan comparison for a typical patient is shown in the figure, which is a cumulative dose mass histogram (DMH) depicting the mass of normal lung tissue receiving at least a certain dose. The lung DMHs shown are for three plans: (A) Free-breathing; (B) Target immobilization only; (C) Target immobilization plus reduced lung density. It is observed that the mass of lung tissue treated to high doses is less for cases B and C. For example, the mass receiving >24.5 Gy is 235 g for free-breathing (A), 180 g for immobilization alone (B) and 150 g for DIBH (C). For this patient, target immobilization alone would allow a dose escalation from 75.6 Gy to 95 Gy for the same level of NTCP as the free-breathing plan, the added effect of reduced density would then enable a further escalation from 95 Gy to 110 Gy. For this study, the measured tumor motions ranged from 5 to 12 mm for free-breathing and 1 to 3 mm for DIBH. For the example shown in the figure the tumor motion was 10 mm and DIBH reproducibility was 3 mm. Additional details on the DIBH technique and results for a group of patients will be presented. Conclusions: Compared to conventional free-breathing treatment the DIBH technique benefits from reduced margins, as a result of the suppressed target motion, as well as a decreased lung density - both contribute to moving normal lung tissue out of the high dose region. With more normal lung tissue below the threshold for l ocal damage, the possibility for dose escalation is improved