Proceedings of the 52nd Annual ASTRO Meeting S815

Results: Both RA and H-RA plans had a better conformity compared with CRT, with an improvement of CI by 20% (p\0.001).The dosimetric quality of H-RA plans is the best. Only one H-RA plan failed to meet the departmental criteria (2 for RA and CRTplans). It produced the lowest V20 (volume of lung-PTV receiving $20Gy) and mean lung dose (MLD), without jeopardizing thelow-dose lung volume (V5). The mean V20 was 23.04%, 1.75% lower than CRT (p \ 0.001) and 3.22% lower than RA (p =0.001); MLD (lung-PTV) was 13.97Gy, 0.55Gy lower than CRT (p = 0.004) and 0.53Gy lower than RA (p = 0.006). Themean V5 was similar in H-RA and CRT plans (p = 0.17) and both were significantly lower than RA plans (p \ 0.01). Themean maximum spinal cord dose of H-RA plans was 34.48Gy, which was the lowest among the three techniques (3.98Gy lowerthan CRT (p \ 0.001) and 1.94Gy lower than RA (p = 0.13). The H-RA plans yielded the lowest NTCP of lung. Though the av-erage MU per fraction was higher for both RA and H-RA compared with CRT plans (an increase of 62.5 MU (p = 0.003) and 52.4MU (p\ 0.001) respectively), the average treatment time of H-RA was 33 seconds faster compared with CRT plans (p = 0.013).

Conclusions: The Hybrid-RapidArc RT technique utilizing two arcs with static fields was capable of producing superior dose dis-tribution without compromising the treatment time for treating advanced stage non-small cell lung cancer compared to CRT andRA technique.

Author Disclosure: O.S.H. Chan, None; A.W.M. Hung, None; M.C.H. Lee, None; A.T. Chang, None; R.M.W. Yeung, None;A.W.M. Lee, None.

3366 Risk of Distant-to-target Dose Deposition using an Intensity-modulated Volumetric Arc Therapy Planning

and Delivery Concept (RapidArc)

P. H. Cooper, J. Tanyi, W. Laub, M. Fuss

OHSU, Portland, OR

Purpose/Objective(s): During review of clinical intensity-modulated volumetric arc therapy plans (RapidArc, Varian MedicalSystems, Palo Alto, CA), islands of radiation dose were noted cranial and/or caudal to the target region. Investigation into thecauses of this observation revealed use of non-zero multi-leaf collimator (MLC) rotation, as the source of this dose deposition.

Materials/Methods: In 16 of 23 RapidArc plans, a distant-to-target focus of dose deposition was observed. Maximum distant-to-target doses ranged from 6.2-34.0% (mean 16.4%) of prescribed dose (PD). The average distance to the most cranial or caudalaspect of the PTV was 28 mm (range, 14 to 50 mm).

Results: A discrete focus of distant-to-target dose deposition was observed in 12 of 18 cases when a collimator angle of 45 degreewas used. Here the mean percentage of PD was 17.2% (range, 6.2 to 34.0%). The percentage of PD located outside the target wascorrelated with length of rotation or time during which a closed MLC leaf pair aligned with the arc rotation axis. A smeared out,sickle-shaped area of dose deposition distant to the PTV was observed with use of collimator angles smaller than 45 degree. Owingto dose blurring over a larger volume, peak doses were lower in those scenarios with mean doses measured of 13.9% of PD (range,9.2 to 19.9%). Dosimetric verification revealed at least a 99.1% agreement between measurement and calculation based on a 3 mm/3% gamma index criterion. The cause of this phenomenon is a tendency of the treatment planning software to ‘‘park’’ the gap be-tween opposing MLC leaves in the upper left and lower right inner corners of the primary collimator rather than under the primarycollimator jaws. Use of a non-zero collimator angle for dynamic arc planning may align these two corners on or close to the cranio-caudal rotational axis of the linear accelerator gantry. Consequently, dose leakage from the leaf gap is deposited in a discrete regiondistant from the target for a variable proportion of the rotational delivery.

Conclusions: Preliminary strategies to avoid this phenomenon include collimator rotation smaller or larger than 45 degrees. How-ever, even at MLC rotation angles of 25 degrees, distant-to-target dose deposition has been observed, albeit to a lesser amount. Ourpreliminary recommendation for RapidArc treatment planning and delivery is review of MLC motion patterns for all plans, avoid-ance of collimator rotation angles of 45 degrees, and careful surveillance for distant-to-target dose deposition. As of this submis-sion, the clinical relevance of this observation remains unclear.

Author Disclosure: P.H. Cooper, None; J. Tanyi, None; W. Laub, Computerized Medical Systems, F. Consultant/Advisory Board;M. Fuss, Varian Medical Systems, B. Research Grant; Varian Medical Systems, D. Speakers Bureau/Honoraria; BrainLAB, D.Speakers Bureau/Honoraria.

3367 Robotic Radiosurgery for Inoperable Patients with Peripheral Stage IA Non-small Cell Lung Cancer:

Local Control and Survival using 5-mm Margins

B. S. Gill1, S. Vahdat2, S. Suy2, S. Lei2, Y. Xia2, F. Banovac2, E. D. Anderson2, S. P. Collins2, A. Dritschilo2, B. T. Collins2

1The George Washington University, School of Medicine, Washington, DC, 2Georgetown University Hospital, Washington, DC

Purpose/Objective(s): To report the clinical efficacy of robotic radiosurgery for inoperable patients with peripheral stage IA non-small cell lung cancer (NSCLC) using 5 mm margins and to determine the V15 values (volume of lung receiving 15 Gy or more) forpatients with 5 mm margins and standard RTOG margins.

Materials/Methods: Twenty-four inoperable patients with biopsy-proven peripheral clinical stage IA NSCLC were enrolled inthis single institution study. All patients had three-to-five gold fiducial markers placed in or near tumors to serve as targeting ref-erences for image guidance. Gross tumor volumes (GTvs.) were contoured using lung windows; the margins were expanded by 5mm to establish the planning treatment volume (PTV). Non-isocentric treatment plans were designed to deliver 42-60 Gy in 3 equalfractions (BED Gy10 .100). Clinical examinations and PET/CT imaging were completed at 6-month intervals following treatment.The V15 values were calculated using 5-mm margins and compared to the V15 values calculated using standard RTOG margins(5 mm in the axial plane and 10 mm in the longitudinal plane).

Results: Tumor diameter ranged from 1.4 to 3.0 cm with a mean maximum tumor diameter of 2.2 cm (range, 1.4-3.0 cm) and a meanGTV of 10 cc (range, 1-24 cc). These 24 patients were accrued over a 4-year interval and followed for a minimum of 1 year. The meandose was 51 Gy delivered to the PTV in 3 fractions over 5- to 11-days. The mean calculated V15 value was 202 cc for the 5 mmmargin and 237 cc for the RTOG margin. With follow-up of 38.7 months (median) the 3-year Kaplan-Meier local control and overallsurvival estimates for the group were 96% and 79%, respectively. Five patients presenting with severe emphysema (baseline postbronchodilator FEV1\40% predicted) died of progressive lung dysfunction at 7, 9, 10, 18 and 25 months.

S816 I. J. Radiation Oncology d Biology d Physics Volume 78, Number 3, Supplement, 2010

Conclusions: Robotic radiosurgery with image guidance offers a precise effective treatment for inoperable peripheral stage IANSCLC. Limiting the margin to 5 mm decreased the V15 by an average of 35 cc (range, 15-60 cc) while preserving local controlrates comparable to published studies using larger margins. Sparing lung tissue in this group of patients is important since deathswere uniformly the result of progressive lung dysfunction.

Author Disclosure: B.S. Gill, None; S. Vahdat, None; S. Suy, None; S. Lei, None; Y. Xia, None; F. Banovac, None; E.D. Ander-son, Accuray, D. Speakers Bureau/Honoraria; S.P. Collins, Accuray, B. Research Grant; A. Dritschilo, None; B.T. Collins, Accu-ray, D. Speakers Bureau/Honoraria.

3368 Dosimetric Study of Intensity Modulated Arc Therapy for Total Marrow Irradiation

C. Han, T. E. Schultheiss, J. Y. C. Wong

City of Hope National Medical Center, Duarte, CA

Purpose/Objective(s): In recent years, helical tomotherapy has been used for total bone marrow irradiation for sparing ofcritical organs. In this study, we evaluated the efficacy of intensity-modulated arc therapy (IMAT) for total bone marrow ir-radiation.

Materials/Methods: Three patients who received total marrow irradiation with helical tomotherapy, two females and one male,were selected for this dosimetric study. Each patient had one CT scan with normal breathing from the skull to the mid-thigh, andtwo CT scans at the end of the expiration and inspiration phases. Contours for target volumes and normal organs were drawn to takeinto account breathing effects. The target volumes included the bone volume from the skull to the mid-thigh, lymph nodes, andspleen. 12 Gy in 8 fractions was prescribed to the target volumes. Varian RapidArc IMAT plans were generated with the sametarget volumes as the helical tomotherapy plans. For each patient, 23 normal structures were included in plan optimization, includ-ing all the major organs in the body. In helical tomotherapy plans, 5 cm jaw width was used. For IMAT planning, eight arc fieldswere used for each patient to cover the entire target volumes, with the isocenters aligned along the superior-inferior axis. Both thehelical tomotherapy and IMAT plans were normalized to cover 85% of the bone volume by 12 Gy.

Results: The average values of the mean doses to the normal organs were 6.0 Gy, 6.0 Gy, and 5.8 Gy in helical tomotherapy plans for thethree patients; they were 4.8 Gy, 5.4 Gy, and 5.2 Gy, respectively, in IMAT plans. Of the 63 organs that could be compared pairwise, 43had lower means using IMAT and 17 had lower means using tomotherapy. Of the 40 organ pairs whose mean doses differed by more than10%, 33 had lower means in IMAT plans. D10 (the minimum dose level covering 10% of the volume) for the bone volume were 13.5 Gy,13.1 Gy, and 13.5 Gy in helical tomotherapy plans for the three patients, respectively; they were 13.7 Gy, 14.4 Gy, and 14.3 Gy in IMATplans. Comparable dose coverage was achieved by IMAT plans for other target volumes. The average beam-on time was 19.4 minutes forhelical tomotherapy and approximately 16 minutes for IMAT. Planning time excluding contouring was approximately 4h for eachmethod. The average monitor unit (MU) in helical tomotherapy plans was 17,207 MU; it was 4,916 MU in IMAT plans.

Conclusions: Compared to helical tomotherapy plans, IMAT plans showed consistent improvement in sparing of normal organswhile delivering a somewhat less homogeneous dose to the target volumes for total bone marrow irradiation. This study indicatesthat IMAT can be used effectively for total bone marrow irradiation. With either treatment lower extremities are most efficientlytreated with conventional open fields.This work supported by a research grant from Varian Medical Systems, Inc.

Author Disclosure: C. Han, None; T.E. Schultheiss, None; J.Y.C. Wong, None.

3369 Volumetric Modulated Arc Radiotherapy for Nasopharyngeal Carcinoma: A Dosimetric Comparison with

Tomotherapy and Step-and-shoot IMRT

S. Lu1, C. Wang1,2, L. Chen1, J. Lee3, S. Wen1, J. Wu1, Y. Lin4, J. Cheng1

1Division of Radiation Oncology, Departments of Oncology, National Taiwan University Hospital, Taipei, Taiwan, 2Institute ofElectrical Engineering, National Taiwan University, Taipei, Taiwan, 3National Yang-Ming University, Taipei, Taiwan,4Cyberknife Center, National Taiwan University Hospital, Taipei, Taiwan

Purpose/Objective(s): Volumetric modulated arc therapy (VMAT) has been recently investigated in prostate cancer therapy be-cause of its highly conformal dose distribution with shorter treatment time. The goal of this study was to compare VMAT withhelical tomotherapy and step-and-shoot intensity-modulated radiation therapy (IMRT) for nasopharyngeal carcinoma (NPC) pa-tients with regard to the sparing effect on organs at risk (OARs) and efficiency of delivery.

Materials/Methods: Seventeen patients with NPC treated by helical tomotherapy were re-planned by VMAT (two arcs) andIMRT (seven to nine fields) for dosimetric comparison. The treatment planning used for VMAT and IMRT was Pinnacle3�-ver-sion 9.0. A dosimetric comparisons between VMAT, tomotherapy, and IMRT plans were analyzed to evaluate (1) coverage ofplanning target volume (PTV), (2) sparing of OARs, (3) delivery time, and (4) monitor units (MU).

Results: The VMAT, tomotherapy, and IMRT plans had similar PTV coverage with an average of 96%. The homogeneity indicesof VMAT (1.063) and tomotherapy (1.058) were better than IMRT plans (1.072, p \0.001). There was no significant differencebetween VMAT and tomotherapy in homogeneity. Tomotherapy plans provided a better conformity index (1.17) than VMAT(1.26, p = 0.014) and IMRT (1.36, p = 0.016). The average maximum doses to spinal cord/brainstem for VMAT, tomotherapy,and IMRT were 31.4/ 49.0, 31.6/ 48.7, and 35.7/52.4, respectively. When compared with IMRT, VMAT and tomotherapy hada better sparing effect on brainstem and spinal cord (p \ 0.05). The effect of parotid sparing was similar between VMAT(mean = 26.5 ± 3.3Gy) and tomotherapy (mean = 27.4 ± 2.0Gy), but better than IMRT (mean = 31.5 ± 2.4Gy, p \ 0.001).The average delivery times for VMAT, tomotherapy, IMRT were 5.5 ± 1.1, 9.3 ± 0.8, and 9.1 ± 2.0 minutes. The monitor unitsper fraction for VMAT were much lower than for tomotherapy: 666.4 ± 45.7, and 8133.9 ± 635.6, respectively.

Conclusions: There were no significant differences between VMAT, tomotherapy, and IMRT in the coverage of PTV; however,VMAT provided better sparing of normal tissue, homogeneity, and conformity than IMRT, and shorter delivery time and fewermonitor units than tomotherapy.

Author Disclosure: S. Lu, None; C. Wang, None; L. Chen, None; J. Lee, None; S. Wen, None; J. Wu, None; Y. Lin, None;J. Cheng, None.