2009 ACMP Meeting – Young Investigators' Sympympymposium Abstracacts

Method and Materials 7 volunteers and 5 lung cancer patients underwent a 5-minute MRI scan in the sagittal plane to acquire dynamic MRI (dMRI) of lung motion, from which motion trajectories of the tumor (for patients) or a pulmonary vessel (for volunteers) were determined. A MATLAB program was written to simulate cine-mode 4DCT by segmenting/resorting dMRI. Image motion phantoms were created by moving a round-disk (mimicking tumor) with dMRI-determined trajectories for all subjects. Simulated 4DCT (sCT) were generated from phantom images, and also from dMRI for patients. Internal target area (ITA, 2D counterpart of ITV) in the gated window was determined from both sCT and dMRI in each of the phantom and patient studies, from which the area (ITAGW), major axis (L1) and minor axis (L2) were determined and compared. Similarity between the two ITAs was also calculated.


Conclusion
4DCT may underestimate the gated window ITV. An additional margin to account for the breathing variation may be necessary in determining gated window PTV.

Investigation of the dosimetric consequences based on imaging used for conventional, gated, and tracking radiotherapy of mobile tumors
Teboh Roland, a Y Liu, N Papanikolaou The University of Texas Health Science Center at San Antonio, San Antonio, TX, USA.

Purpose
Investigate the dosimetric consequences based on imaging used for conventional, gated, and tracking radiotherapy of mobile tumors.

Method and Materials
Data from two patients previously treated for lung cancer with approximately 1 cm tumor extent of motion was used. Both 3D and 4D dose distributions were computed, with 4D accounting for organ motion and based on a 4DCT image set while 3D dose was based on a static anatomy. The 4D dose was derived from multiple 3D plans corresponding to various phases of the respiratory cycle via a validated non-rigid deformable image registration algorithm. We compared the 3D versus 4D plan predicted lung volume irradiated by at least 20Gy (V20), mean lung dose (MLD), isocenter point dose (IPD), and the target generalized equivalent uniform dose (GTV-gEUD and PTV-gEUD) for conformal radiotherapy (no margins to account for motion -3DCRT; with 4DCT derived margins to account for motion -4D Static), gated, and tracking radiotherapy.

Conclusion
Although we observed some trend in the dosimetric parameters considered (for example, the discrepancies across all parameters for the 4D static and gating techniques were within 2%), further studies involving varied tumor characteristics are required for any concrete conclusions.

MO-F-Sandler-03
Intensity-modulated proton planning for ocular tumor using human anatomy dose algorithm and preliminary comparison with IMRT planning

Purpose
We aim to provide accurate dose calculations for ocular tumor and adjacent critical organs using intensity-modulated proton therapy (IMPT) using a human anatomy-based Monte Carlo model. Dose is simulated using the Monte Carlo code MCNPX and compared to standard photon IMRT planning using Pinnacle3 TPS.

Results
The minimum, mean and maximum PTV doses were 53.07, 62.23 and 66.32 Gy, respectively. The dose heterogeneity index (D 1% /D 99% ) was 1.14%. Accounting for heterogeneities from tissue and dose distributions, the SFED was 4% less than Park's value.

Conclusions
The SFED (1) with EUD is suitable for describing and comparing SBRT prescription schemes as it incorporates the strengths of the linear quadratic model for doses near the shoulder region and the multi-target model for large doses that fall on the exponential part of the cell survival curve, giving an accurate description of the equivalent potency of SBRT. Use of EUD is more biologically relevant than use of prescribed dose, since it falls between minimum and mean dose on the DVH.

Purpose
Optimum treatment efficacy requires that setup uncertainties be determined for each institution, treatment site, and technique. This work investigates how the use of volumetric image guidance affects tumor targeting accuracy and tumor volume margin requirements in SBRT lung patients. These results are compared to the RTOG 0618 SBRT lung protocol requirements.

Method and Materials
Setup uncertainties and margins were established for forty-five patients treated with stereotactic lung radiotherapy using the Varian OBI CBCT imager for daily target localization. At each treatment fraction, at least three CBCTs were acquired to: 1) assess the initial required shifts, 2) verify the patient position before treatment, and 3) assess the patient position after treatment.

Results
The group margins were calculated according to the Van Herk formalism with the intent to give 95% of the prescribed dose to 90% of the patient population. With laser alignment, an average margin of 20.6 mm in each direction is required. After CBCT localization, the average margin requirement in each direction is reduced to 3.3 mm. If the intrafraction tumor motion is taken into account, this margin is increased to 4.0 mm.

Conclusion
The amplitude of the 95%-90% margin (3-5 mm, depending on direction) is consistent with the RTOG requirements of 5-10 mm. Larger margins might be required if higher dose or population coverages are intended.