S-Frame Versus Dual Vacuum: A Comparison of Immobilization Techniques for Upper Lung Tumors With Stereotactic Body Radiation Therapy

Purpose/Objective(s): To compare the impact of SBRT immobilization strategies (S-frame/Dual vacuum) on setup and reproducibility for upper lung tumors. Materials/Methods: Our institutional prospective SBRT database was searched for patients with upper lung tumors, defined as an internal tumor volume superior to T5 vertebra. Initially the full-length dual vacuum was used for all lung SBRT patients. For upper lung tumors, an extended, thermoplastic S-frame was recently added as an option. At simulation, a free-breathing 4-D computed tomography was performed without respiratory gating or breathing control. For each treatment, patients were first set up to isocenter with in-room lasers and skin tattoos. Subsequent shifts from initial couch position were made using cone-beam CT (CBCT) to correct for tumor position. If shifts greater than 4 mm were required, they were applied and a new CBCT would be obtained. Couch shifts were analyzed as square root of sum of squares in x, y, and z-directions. Kaplan-Meier, Fisher’s exact and equal variance t-tests were used. Results: Median follow-up was 8.7 months (quartiles, 4.4-19.1). Between April 2008 and Jan 2012, 88 patients with 93 upper lung tumors (21 central, 72 peripheral) were identified. Median age was 71.8 years. 64 tumors were non-small cell lung cancer (54 primary, 10 recurrent), 29 metastatic from other sites. Lesion sizes were: 3cm (11%). 17 tumors were immobilized with S-frame; 76 with dual vacuum. Dose schemes were: 48Gy/4 fractions (28%), 50Gy/5 fractions (24%), and 54Gy/3 fractions (46%). A total of 347 fractions (65 S-frame, 282 dual vacuum) were delivered. First couch shifts were statistically better with S-frame than dual vacuum (see Table). The first couch shift by CBCT exceeded the 4-mm tolerance in 33 (51%) fractions in S-frame, vs. 217 (77%) in dual vacuum (Fisher’s p < 0.0001). Correction by 1st couch shifts was usually successful, as 2nd couch shifts were similarly small regardless of immobilizing device; 3rd shifts were rarely needed. For all patients, 2 and 3-year local control rates were 94% and 81%, respectively; median has not been reached yet. There is no difference