Proceedings of the 49th Annual ASTRO Meeting S659

2826 Interactive Image Guided Peripheral Brachytherapy for the Boost Dose in Breast Irradiation

M. J. Rivard1, R. J. Bricault2, C. S. Melhus1, J. R. Hiatt3, P. Sioshansi2, D. E. Wazer1,4

1Tufts–New England Medical Center, Boston, MA, 2Advanced Radiation Therapy, LLC, Billerica, MA, 3Rhode Island Hospital,Providence, RI, 4Rhode Island Hospital, Providence, RI

Purpose/Objective(s): The AccuBoost system (Advanced Radiation Therapy, Billerica, MA) applies breast brachytherapy with-out invasive catheters, and with mammographic image guidance using stereotactic localization of markers implanted during lump-ectomy. This high-resolution imaging permits accurate localization of the region to be treated and assists with radiotherapy beampositioning. A combination of parallel-opposed HDR 192Ir beams are directed into the breast towards maximizing target dose uni-formity and minimizing dose to skin and other healthy tissues. Pre-clinical studies examining dosimetric characterization andsystem commissioning for clinical use are presented.

Materials/Methods: The AccuBoost system is FDA-approved, and includes a fully-functional mammography system (modelSenographe 800T by GE Healthcare, Piscataway, NJ), a computed radiography (CR) image analysis system (model CR 825 byKodak, New Haven, CT) using a phosphor plate, three applicators with inner diameter ranging from 5–7 cm, and custommammography paddles for precision applicator positioning on a compressed breast. The applicators have a single 6F catheterlead which is compatible with Nucletron and Varian HDR 192Ir remote afterloading systems. Patients treatments entail slight com-pression (3–7 cm) of the breast, paddle positioning, mammographic imaging and CR plate exposure, image analysis and possiblepaddle repositioning, treatment delivery calculations with the appropriately sized applicator, and parallel-opposed irradiation. To-wards benchmarking this novel treatment modality, localization accuracy based on image-guidance of the paddle/applicator assem-bly was assessed with a breast phantom (model 051 by CIRS, Norfolk, VA) containing radio-opaque markers under a variety ofclinically-relevant conditions. Measurements (ionization chambers and GafChromic EBT film) were performed and compared toMonte Carlo (MCNP5) simulations of dose distributions as a function of depth, position, and for various breast compression thick-nesses (Fig.).

Results: Under all conditions, agreement of MC simulations with ion chamber results were within 2% for dose falloff and within6% for dose profiles measured using film. For 4 orthogonal beams, calculated skin: target dose ratio varied between 0.5–1.1; morefavorable results were observed for thinner compressed breast thicknesses. Marker localization within 1 mm was readily achiev-able, and imaging with 192Ir was examined.

Conclusions: Under image-guidance, the AccuBoost system can apply a conformal and uniform breast boost dose. MC simula-tions readily characterized the clinical environment as shown through agreement with measured results.

Author Disclosure: M.J. Rivard, None; R.J. Bricault, employment, A. Employment; C.S. Melhus, None; J.R. Hiatt, None; P. Siosh-ansi, employment, A. Employment; D.E. Wazer, None.

2827 Evaluation of an Adaptive Diffusion Smoothing Technique to Improve Delivery Efficiency in Accelerated

Partial Breast IMRT

M. M. Matuszak, R. B. Marsh, E. W. Larsen, B. A. Fraass, L. J. Pierce, J. M. Moran

University of Michigan, Ann Arbor, MI

Purpose/Objective(s): Accelerated partial breast irradiation (APBI) is under investigation at many institutions for early-stagebreast cancer treatment. While previous studies have shown that a strict IMRT protocol can improve target coverage and reducenormal tissue dose compared to 3D conformal therapy, there are disadvantages of the increased delivery time and monitor unitsrequired for IMRT. The purpose of this study is to evaluate the use of an adaptive diffusion smoothing (ADS) method duringoptimization to improve delivery efficiency while preserving target coverage and normal tissue sparing.

Materials/Methods: A diffusion-based method to reduce IMRT beam complexity has been developed at our institution. This ADSprocedure penalizes areas of non-essential modulation based on computer-defined coefficients that dictate the amount of smoothingapplied in each beamlet. To evaluate the impact of applying ADS during optimization of APBI cases, IMRT plans were generatedretrospectively for 10 patients treated on an IRB-approved protocol at our institution. Under this protocol, 38.5 Gy is delivered in 10fractions with strict homogeneity and normal tissue dose objectives. After optimization with the protocol objectives to establisha baseline plan, three plans were optimized with ADS. First, the baseline objectives were constrained and ADS was applied to de-termine the delivery efficiency improvement possible with no loss in plan quality. Next, the minimum target dose was relaxed by 3and 5% to determine the further smoothing possible with small concessions in homogeneity. The plan metrics, beam complexity,and delivery efficiency for all plans were analyzed.

Results: The MU required for all plans are shown in the Table relative to the baseline plan. The MU reduction possible when ap-plying ADS was 24.6 ± 7.1% with no loss in plan quality, 34.2 ± 8.3% with a 3% relaxation in minimum target dose, and 39.4 ±7.7% with a 5% relaxation. All differences observed in the plan metrics were negligible with the exception of the minimum targetdose in the final two ADS plans.

Conclusions: The use of ADS during optimization of APBI cases allows the full dosimetric benefit of intensity modulation to berealized without the substantial loss in delivery efficiency that may be observed with conventional IMRT. In 10 patients, the mean