Compressed air tissue expanders (CATEs) consist of a silicon shell containing a metallic CO2 reservoir, surgically placed in the chest wall post-mastectomy. CATEs pose significant challenges for RT: The high density reservoir causes artifacts on the planning CT, which encumber structure definition and cause misrepresentation of density information, in turn affecting dose calculation. This study describes a method to model the CATE in a commercial treatment planning system (TPS), and discusses the limitations of different dose calculation algorithms (DCAs) in and around the device. An understanding of DCA accuracy near the CATE is critical for assessing individual plan quality, appropriateness of DCA and planning technique, and even the decision to use a CATE in an RT setting. A CATE model was created in a commercial TPS. The CATE was imaged using optimal CT geometry and technique. Individual components were contoured and dimensions were verified against manufacturer specifications. The model was available for registration with a patient CT in the TPS. Assigned densities of the model were optimized by comparing measured and calculated transmission through the CATE. Transmission was measured with radiochromic film in various geometries. Dose was calculated using two commercially available DCAs: a convolution-based algorithm (CBA) and an explicit linear Boltzman transport equation solver algorithm (LBTEA). Doses were compared using profile and gamma analyses. Clinical impact was evaluated using CT data from 3 patients with CATEs. The CATE model was registered to each patient CT, and 3DRT plans were calculated using both DCAs. Clinically significant DVHs were analyzed. For direct transmission through the CATE, both DCAs achieved greater than 99% gamma pass rate. However, for the region 0-1 cm adjacent to the CATE, gamma pass rate was greater than 98% for the LBTEA, but near 0% for the CBA. Compared to the LBTEA, the CBA overestimated mean dose in the CO2 reservoir and the air cavity by 5-7% and 10-13%, respectively. The CBA underestimated mean dose in the reservoir’s dose shadow and the adjacent chest wall by 0.5-5%. Changes in max dose were more variable and patient-specific. Approaching a CATE in an RT setting, clinicians must first obtain accurate patient density information. The CATE model described here is one practical method. Second, clinicians must understand the accuracy of their DCA near the CATE to evaluate plan quality. This work suggests that both the CBA and LBTEA accurately predicted beam transmission through the CATE, but the CBA overestimates dose within 1 cm lateral of the CATE. Finally, clinicians must decide if and how an acceptable plan can be delivered with the CATE in place. The decrease in target coverage due to attenuation by the CATE may require optimization techniques (such as field-in-field). These techniques should be used with careful consideration of the DCA’s uncertainty.

Purpose: To describe the development of a knowledge‐based treatment planning model for lung cancer patients treated with SBRT, and to evaluate the model performance and applicability to different planning techniques and tumor locations. Methods: 105 lung SBRT plans previously treated at our institution were included in the development of the model using Varian's RapidPlan DVH estimation algorithm. The model was trained with a combination of IMRT, VMAT, and 3D–CRT techniques. Tumor locations encompassed lesions located centrally vs peripherally (43:62), upper vs lower (62:43), and anterior vs posterior lobes (60:45). The model performance was validated with 25 cases independent of the training set, for both IMRT and VMAT. Model generated plans were created with only one optimization and no planner intervention. The original, general model was also divided into four separate models according to tumor location. The model was also applied using different beam templates to further improve workflow. Dose differences to targets and organs‐at‐risk were evaluated. Results: IMRT and VMAT RapidPlan generated plans were comparable to clinical plans with respect to target coverage and several OARs. Spinal cord dose was lowered in the model‐based plans by 1Gy compared to the clinical plans, p=0.008. Splitting the model according to tumor location resulted in insignificant differences in DVH estimation. The peripheral model decreased esophagus dose to the central lesions by 0.5Gy compared to the original model, p=0.025, and the posterior model increased dose to the spinal cord by 1Gy compared to the anterior model, p=0.001. All template beam plans met OAR criteria, with 1Gy increases noted in maximum heart dose for the 9‐field plans, p=0.04. Conclusion: A RapidPlan knowledge‐based model for lung SBRT produces comparable results to clinical plans, with increased consistency and greater efficiency. The model encompasses both IMRT and VMAT techniques, differing tumor locations, and beam arrangements. Research supported in part by a grant from Varian Medical Systems, Palo Alto CA.

We report on a retrospective evaluation of 4D-CT images and plans for 7 patients with pancreatic cancer. The aims of this study are to analyze the extent of intrafraction motion of the pancreas, and to evaluate the dosimetric consequences of motion on the planned doses to the target and healthy tissues. The 4DCTs were acquired and sorted to generate respiratory correlated datasets at multiple phases of the breathing cycle. An internal target volume (ITV) was derived from a composite union of the GTvs. on each of the datasets. The PTV was formed by a 3 mm uniform expansion of the ITV except in the vicinity of adjacent normal organs, where no margin was used. Optimization was performed on the free-breathing (FB) scan and doses were mapped onto each phase. A weighted composite plan containing doses for all breathing phases was created and mapped onto the FB CT scan. Individual phase doses were weighted according to the 60%-10%-20%-10% ratio for Exh-Mid-Inh-Inh-Mid-Exh1. Also, a conventional, non-4D plan was designed on the exhale phase using a uniform 1 cm GTV-to-PTV expansion. The dose prescription was typically 30 or 35 Gy delivered in 5 IMRT fractions. Excursion of the pancreas center of mass (COM) was 0.48 ± 0.27 cm (μ ± σ) with the mean ranging between 0.2 and 0.93 cm. Deformation effects were found to be clearly evident resulting in some regions of the tumor moving more than the COM. For one patient the inferior aspect of the tumor moved 1.2 cm in SI direction while the COM moved 0.93 cm. Analysis of doses to the target indicated that a 1 cm margin expansion of the GTV-to-CTV in the FB-based, non-4D plan, resulted in approximately equivalent coverage relative to the composite 4D plan. However, significant volume and dose differences were sometimes observed between the FB-based and 4D plans for the organs at risk. Concerning volume differences, the volume of the liver on the FB scan was anywhere from 7 to 14% lower than on the phase-specific CT images. This resulted in up to 20% higher max. doses to liver as compared to the original FB-based plan, when the plan was mapped onto phase-specific images. Mean liver doses were also underestimated when on the FB scan by, on average, 25% vs. the composite 4D plan. Relative to plans on the exhale phase, the mean liver dose was up to 28% greater than that in the composite 4D plan (averaged over all phases) suggesting that the exhale plan provides a worst case estimate of liver dose. Analysis of 4DCTs of patients with pancreatic cancer treated with SBRT showed significant intrafraction motion and deformation effects. A non-4D, FB-based plan using a 1 cm PTV margin achieved adequate target dose coverage relative to a composite 4D plan. Significant dose differences to surrounding healthy tissues were observed.