S136 I. J. Radiation Oncology d Biology d Physics Volume 78, Number 3, Supplement, 2010

Results: With the1 min. irradiation, the neutron yield from brass is about a factor of 3 lower than that of W, and the decay of ac-tivities is about 10 times faster. For the second simulation, the neutron activity builds up slowly in brass, from about 7x103 Bq/cm3

at the end of week one to a maximum of 104 Bq/cm3 at the end of week 7. For W, the neutron activity builds up slowly from about5.5x106 Bq/cm3 at the end of week one, to about 9x106 Bq/cm3 (0.24 mCi) at the end of week 5. The neutron activities at thebeginning of the week (during irradiation) are due to nuclides with atomic numbers in the range Z = 8 - 78. At the end of theweek, the neutron activities are mostly due to nuclides in the range Z = 60-78, and a smaller contribution from nuclides in the rangeZ = 20-60.

Conclusions: Our simulation showed that brass results in much lower neutron production than W. Although neutron activityslowly builds up in W, the activity is only about 0.24 mCi after 5 weeks at the surface of the metal. At the patient level, the activitywill be smaller due to inverse square law. Our next step is to simulate a realistic treatment schedule with a commercial MLC to moreaccurately predict the neutron activities.

Author Disclosure: V.P. Moskvin, None; C. Cheng, None; D. Nichihporov, None; I. Das, None.

290 Plan-specific RBE Calculation using Monte Carlo for Proton Therapy

W. Luo1, J. Li2, J. Fan2, C. Ma2

1Univerisity of Kentucky, Lexington, KY, 2Fox Chase Cancer Center, Philadelphia, PA

Purpose/Objective(s): A generic and constant RBE of 1.1 has been widely used for dose correction in proton therapy, which is notaccurate. In this study, an accurate and applicable RBE calculation for specific patient plans was proposed based on the MonteCarlo simulation.

Materials/Methods: The RBE calculation in Monte Carlo was based on the relationship between RBE and linear energytransfer (LET),

RBE�D;L; a0; k; a; b

�¼��

a2 þ 4bD�a0 þ kLþ bD

��1=2�a�.�

2bD�

[1]

The parameters D,L,a0,k,a,b are determined by experiments and L is calculated using the Bethe-Bloch formula. Monte Carlosimulations for proton dose calculation were implemented with MCDOSE combined with GEANT3. Scanning proton pencilbeams (1x1cm2) were simulated for this study. An algorithm has been developed to calculate the energy spectra and weights cor-responding to the desired SOBPs. Proton therapy (PT) without intensity modulation was performed using those beams. The protonbeams can be further optimally weighted to generate intensity-modulated proton therapy (IMPT). The gradient search algorithmwas used in the optimization process to generate IMPT plans. The RBE values were also estimated by the ratios of various doseparameters between RBE and without RBE calculation.

Results: The proton plans were generated based on 10 real prostate patient’s CT data and structures. The 1-beam, 2-beam, 4-beam and7-beam plans were presented. The corresponding IMPT plans were also produced to show if the RBE would be affected by intensitymodulation. The RBE was calculated specific to each dose parameter. The results showed that all RBE values except the bladder in theone-beam plan were higher than 1.10, ranging from 1.12 to 1.37 for the rectum in the one-beam plan. The one-beam plan had higherRBE for the rectum dose, while the two-lateral beam plan had the lowest RBE of 1.14 for the rectum. The one-beam plan gave the lowestbladder RBE of 1.10 although its bladder dose was the highest. The RBE for prostate mean dose and whole body dose was comparablefor all three plans. The RBEs for those dose points were comparable between PT and IMPT. We also extended the RBE to the volumepoints (V25-volume of 25 Gy, V65-volume of 65 Gy) and defined the volume ratios as the volume related RBE, or RBEV. The RBEVfor V25 and V65 for the bladder (VB25 and VB65) and rectum (VR25 and VR65) varied a lot from plan to plan. The 4-beam IMPTobtained the highest RBEV of 2.89 for VR65, and the 7-beam plan received the highest RBEV of 1.42 for VB65.

Conclusions: The RBE values are not constant and vary from organ to organ and plan to plan. Although the RBEs for the PTV didnot differ dramatically from 1.10, they appeared in a significant range for critical organs such as the rectum and bladder. Therefore,the RBE values and the RBE-corrected doses should be calculated for each proton plan.[1] J. Wilkens and U. Oelfke, PMB. 49 (2004) 2811-2825.

Author Disclosure: W. Luo, None; J. Li, None; J. Fan, None; C. Ma, None.

291 Radiation Induced Rib Fractures after Hypofractionated Stereotactic Body Radiation Therapy: The Risk

Factors and Dose-Volume Relationship

K. Asai1, Y. Shioyama1, S. Ohga1, T. Nonoshita1, T. Yoshitake1, K. Ohnishi1, K. Terashima1, K. Matsumoto1, H. Hirata2,

H. Honda1

1Department of Clinical Radiology, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan, 2Department ofHealth Sciences, Faculty of Medical Sciences, Kyushu University, Fukuoka, Japan

Purpose/Objective(s): The purpose of this study is to clarify the incidence, the risk factors, and the dose-volume relationship ofradiation induced rib fracture after hypofractionated stereotactic body radiation therapy (SBRT).

Materials/Methods: Between April 2003 and March 2007, 174 consecutive patients treated with hypofractionated SBRT for pri-mary or metastatic lung cancer were reviewed. The inclusion criteria were at least 3 months of follow-up by CT scan and no pre-vious overlapped radiation exposure. Radiation induced rib fractures were defined as rib fractures located in the radiation field,detected by CT scan after treatment. The risk factors considered; age, gender, GTV diameter, chest wall - tumor distance werereviewed and each parameter was divided into two groups. Dose-volume histogram analysis was conducted on ribs receivedover 20 Gy at maximal point dose. The max dose and absolute volume received; $10 Gy, $20 Gy, $30 Gy and $40 Gywere determined for each ribs as the dosimetric parameters. The 3- and 5- year Kaplan-Meier (KM) estimates of rib fracturewere calculated. Each risk factor was assessed by a log-rank test. The optimal cut off value for each dosimetric parameter wasanalyzed through the use of receiver-operating characteristic (ROC) curves. The area under the curve (AUC) values were alsocalculated. To estimate the cumulative risk of fracture, the ribs were divided into two groups according to the cutoff value andcompared by log-rank test.