Materials/Methods: Modulated electron radiotherapy (MERT) evaluation was conducted with a Trilogy 120 leaf MLC for6–20 MeV. To provide a sharp penumbra, modulated beams were delivered with short SSDs (70–85cm). Segment widths (SW)ranging from 7–10cm were configured for delivery and planning, using BEAMnrc MC code with 109 particles, andDOSXYZnrc calculations. Calculations were set with: voxel size 0.2 x 0.2 x 0.1cm3, and photon/electron transport energycutoffs of 0.01MeV/0.521MeV. Dosimetry was performed with film and micro chambers. Calculated and measured data wereanalyzed in MatLab. Once validation of static fields was successfully completed, modulated portals (segmented and dynamic)were configured for treatment and calculations. Optimization for target coverage and OAR sparing was achieved by choosingenergies according to target depth, and SW according to spatial coverage. Intensity for each segment was optimized by MCmethods.

Results: Beam sharpness (penumbra) degraded with: decreasing energy and SW, and increasing SSD. PDD decreasedsignificantly with decreasing SW. The top figure demonstrates excellent calculation/measurement agreement (�3mm). Equaldose profiles were achieved with delivery by static, segmental or dynamic MLC, except in the periphery and deep depth regions,which were lower with static delivery. With segmented delivery, we introduced small (�1.5mm) gaps between segments tohomogenize distributions at prescription depth. The bottom figure demonstrates conformal coverage of a target. The treatmenttime to deliver 5 segments of 3 energies was �90s, including console reprogramming.

Conclusions: This study shows MERT as delivered with existing photon MLC is feasible, and provides conformal beam dosedistributions.

Author Disclosure: E.E. Klein, None; M. Vicic, None; C. Ma, None; J. Li, None; S. Stathakis, None; D.A. Low, None.

2819 GEOS/IMRT Versus Tomotherapy: A Comparative Study

A. Van Esch1, D. P. Huyskens1, S. Bontemps2, M. Devillers2, E. Salamon2

1QA Team in Radiotherapy Physics, Tildonk, Belgium, 2Radiotherapy Department, Hopital Ste-Elisabeth, Namur, Belgium

Purpose/Objective(s): : Although highly conformal treatments can already be obtained with IMRT, until now, the choice ofgantry angles has been left up to the expertise of the user. A new inverse planning based algorithm has been made availablefor the geometric optimisation (GEOS) of the angles of incidence and the number of beams (Varian Medical Systems, PaloAlto). We have evaluated whether these new modules are a valuable asset in the inverse treatment planning process and howthey compare to results obtained with tomotherapy (TomoTherapy Inc., Madison).

Materials/Methods: To compare the overall treatment quality that can be achieved with GEOS/IMRT versus helicaltomotherapy, treatments plans were made with both modalities for a library of 20 patients covering a variety of target volumes(e.g. head & neck, rectum, ethmoid, prostate, esophagus). Tomotherapy planning was performed with the TomoTherapy Hi.ArtSystem. GEOS/IMRT was performed on Eclipse (Gen8) on a 21EX dual energy Clinac (6MV and 18MV); coplanar andnon-coplanar solutions were investigated. Both treatment units are available in the department.

Results: In the comparison of the final treatment plans the following endpoints are quantified:1) dosimetric quality of the treatment plan: target coverage, dose to organs at risk, total body dose2) process efficiency: planning time, time required for pre-treatment QA, treatment time

S670 I. J. Radiation Oncology ● Biology ● Physics Volume 66, Number 3, Supplement, 2006

Conclusions: Highly conformal treatment plans have been obtained with both modalities in most cases. Superior dosimetricresults were obtained with GEOS/IMRT for treatment sites benefiting from non-coplanar beam incidences. Sparing of theorgans at risk was more easily achieved in the head&neck region with tomotherapy. GEOS/IMRT offers a more time efficientsolution on almost all aspects. This may partly be attributed to the novelty of the TomoTherapy solution and the growing painsthat go along with it.

Author Disclosure: A. Van Esch, Varian Medical Systems, B. Research Grant; D.P. Huyskens, Varian Medical Systems, B.Research Grant; S. Bontemps, None; M. Devillers, None; E. Salamon, None.

2820 Hypofractionated Stereotactic Radiation Therapy for Pulmonary Nodules Using a Unique FramelessStereotactic Treatment Technique

R. A. Hsi, B. L. Madsen, H. T. Pham, G. Song, K. R. Badiozamani, H. Parsai, L. J. Esagui, D. G. Thompson,E. Hauptmann

Virginia Mason Medical Center, Seattle, WA

Purpose/Objective(s): To assess the feasibility of hypofractionated stereotactic radiation therapy (SRT) for the treatment ofsolitary pulmonary nodules using a unique frameless stereotactic technique.

Materials/Methods: Ten patients with isolated pulmonary nodules who were not candidates for surgical resection underwenthypofractionated SRT receiving 45–50 Gy in 5 fractions. The diagnosis of malignancy was made either by needle biopsy orhighly suspicious PET scan. Each patient underwent CT guided placement of three gold fiducial markers (1.2 x 3 mm) into thethoracic vertebral bodies at the approximate level of the pulmonary nodule. Markers were introduced using a 15 gauge bonebiopsy needle. Six limited radiation treatment planning CT scans were obtained in succession, each at normal end inspiration.The composite CTV was determined by the fusion of CTVs from each of the six planning CT scans. The PTV was defined asthe CTV plus a 1cm margin in the superior/inferior dimension and a 0.5 cm margin in the anterior/posterior and lateraldimensions. A 5–7 field beam arrangement using 6 MV photons was developed using a commercially available radiationtreatment planning system. Dose was prescribed to cover the PTV with the 95% isodose line. Daily stereotactic set up basedupon spinal fiducial marker positions was performed before each treatment using the Acculoc® stereotactic localization system.Patients were treated with the same breath hold technique used to obtain the treatment planning CT scans.

Results: All patients underwent fiducial marker placement without complication and with no evidence of marker migration. Thecomposite CTV volumes were, on average, 20% larger than the individual CTV volumes for each patient. Each patientcompleted treatment without complication. Average treatment time per fraction was 25 minutes. No acute side effects werenoted in any patient.

Conclusions: Hypofractionated SRT for the treatment of solitary pulmonary nodules using this frameless stereotactic techniqueemploying spinal fiducial markers and patient controlled breath hold is feasible and can be achieved with readily availablecommercial treatment planning and localization systems.

Author Disclosure: R.A. Hsi, None; B.L. Madsen, None; H.T. Pham, None; G. Song, None; K.R. Badiozamani, None; H. Parsai,None; L.J. Esagui, None; D.G. Thompson, None; E. Hauptmann, None.

2821 IMRT for Nasopharyngeal Cancers: Potential Advantages for Using Helical Tomotherapy

C. Yang, S. Narayan, F. Guo, C. Wu, J. Perks, S. Vijayakumar, J. Purdy

University of California Davis Cancer Center, Sacramento, CA

Purpose/Objective(s): Rigorous studies comparing various forms of IMRT such as helical tomotherapy versus conventionalMLC IMRT have not been done. As our institution was one of those that participated in the RTOG-0225 protocol, we are ina position to explore the difference between the two modalities in treating nasopharyngeal cancers.

Materials/Methods: Three patients with locally advanced nasopharyngeal cancers that were entered into the RTOG-0225 studyfrom our institution were selected for this study. For each patient, CTV70, CTV59.4 (divided into upper and lower neck portions)were expanded by 0.5 cm to form correspondent PTVs. The 3 patients were planned on Eclipse TPS (Varian Medical System,Palo Alto, CA) and treated using segmental MLC (SMLC) technique on a Varian Cl2100C. One patient’s image and contourdata were transferred from Eclipse TPS into the Tomotherapy HI-ART TPS (TomoTherapy Inc., Madison, WI) and plannedusing 2.5 cm jaw, 0.3 pitch, and 2.5 initial modulation with the same planning goals in RTOG-0225. Dose volume histogramswere exported from both planning systems and compared using MatLab/Excel.

Results: All three original plans of the SMLC technique met the protocol criteria with minor deviation in PTV dose coverage.The averaged PTV70 planning results are: volume: 577 cc and one SD of 203 cc; V77 (percentage of PTV70 receiving 77 Gy):4.1% and 2.3%; V70: 93.8% and 1.8%; V66.5: 99.1% and 0.6%. The averaged dose received by 50% of the contralateral parotidgland is: 27.7 Gy and one SD of 1.4 Gy. In comparison, Tomotherapy achieved a more homogeneous dose coverage of the PTVswhile reducing dose to the spinal cord and contralateral parotid (Figure 1). Mandible receives a mean dose of 24.1 Gy (Tomo)versus 52.8 Gy (SMLC) and larynx: 22.3 Gy (Tomo) versus 40.5 Gy (SMLC). All other critical normal structures have a betteror comparable DVH in Tomo plan than SMLC one. Additionally, Tomo optimization process takes significantly shorter timethan the SMLC one (8 hours versus 16 hours in this example) to satisfy the protocol requirement.

Conclusions: Helical tomotherapy shows real advantages over SMLC based IMRT in normal tissue sparing andhomogeneous target coverage in nasopharyngeal cancers. While a plan generated on Tomotherapy TPS can meet the moststringent requirement specified in RTOG-0225, several features are still needed including basic planning tools such asbeing able to add new structures (contours) and Room’s-Eye-View 3D dose cloud plan review. Most importantly, aDICOM data export compliant with the Advanced Technology QA Consortium DICOM conformance statement is stillneeded.

S671Proceedings of the 48th Annual ASTRO Meeting