S Qi

Denoised and texture enhanced MVCT to improve soft tissue conspicuity

PURPOSE MVCT images have been used in TomoTherapy treatment to align patients based on bony anatomies but its usefulness for soft tissue registration, delineation, and adaptive radiation therapy is limited due to insignificant photoelectric interaction components and the presence of noise resulting from low detector quantum efficiency of megavoltage x-rays. Algebraic reconstruction with sparsity regularizers as well as local denoising methods has not significantly improved the soft tissue conspicuity. The authors aim to utilize a nonlocal means denoising method and texture enhancement to recover the soft tissue information in MVCT (DeTECT). METHODS A block matching 3D (BM3D) algorithm was adapted to reduce the noise while keeping the texture information of the MVCT images. Following imaging denoising, a saliency map was created to further enhance visual conspicuity of low contrast structures. In this study, BM3D and saliency maps were applied to MVCT images of a CT imaging quality phantom, a head and neck, and four prostate patients. Following these steps, the contrast-to-noise ratios (CNRs) were quantified. RESULTS By applying BM3D denoising and saliency map, postprocessed MVCT images show remarkable improvements in imaging contrast without compromising resolution. For the head and neck patient, the difficult-to-see lymph nodes and vein in the carotid space in the original MVCT image became conspicuous in DeTECT. For the prostate patients, the ambiguous boundary between the bladder and the prostate in the original MVCT was clarified. The CNRs of phantom low contrast inserts were improved from 1.48 and 3.8 to 13.67 and 16.17, respectively. The CNRs of two regions-of-interest were improved from 1.5 and 3.17 to 3.14 and 15.76, respectively, for the head and neck patient. DeTECT also increased the CNR of prostate from 0.13 to 1.46 for the four prostate patients. The results are substantially better than a local denoising method using anisotropic diffusion. CONCLUSIONS The authors showed that it is feasible to extract more soft tissue contrast information from the noisy MVCT images using a nonlocal means 3D block matching method in combination with saliency maps, revealing information that was originally unperceivable to human observers.

Tomotherapy improves local control and changes failure patterns in locally advanced malignant pleural mesothelioma.

PURPOSE The purpose of the study was to determine whether intensity modulated radiation therapy delivered via helical tomotherapy improves local control (LC) after pleurectomy/decortication (P/D) for malignant pleural mesothelioma compared with 3-dimensional conformal radiation therapy (3D-CRT). METHODS AND MATERIALS Forty-five consecutive patients were treated with adjuvant radiation to 45 Gy in 1.8 Gy fractions after P/D between 2006 and 2014; 23 received 3D-CRT, and 22 received tomotherapy. Kaplan-Meier analysis was used to calculate overall survival, time to in-field or local failure (LF), and time to out-of-field failure. The Student t test and Fisher exact test were used to detect between-group differences. RESULTS Median follow-up time was 19.4 months and 12.7 months for the 3D-CRT and tomotherapy groups, respectively. Eighty-two percent of patients had T3/T4 disease, and 64% had positive nodes; 17.4% and 41% of patients in the 3D-CRT and tomotherapy groups had nonepithelioid histology, respectively. Mean planning target volume dose, percentage of planning target volume receiving 100% of the prescription dose, and lung doses were significantly greater with tomotherapy (P < .05), but toxicity rates (including radiation pneumonitis rates) were equivalent. LC was significantly improved with tomotherapy on Kaplan-Meier analysis with outcomes censored at 2 years (P < .05); uncensored, this became a trend (P = .06). Median time to LF was 19 months with tomotherapy and 10.9 months in 3D-CRT (the latter interval being less than the median follow-up in the tomotherapy group). On univariate analysis, treatment modality was the only significant predictor of LC (P < .05). Isolated LF was significantly more frequent with 3D-CRT (P < .05). Conversely, isolated out-of-field failure was significantly more frequent with tomotherapy (P < .05). Overall survival and out-of-field control were not significantly different. CONCLUSION Tomotherapy after P/D for malignant pleural mesothelioma is associated with improved target coverage that translates into improved LC compared with 3D-CRT. This is related to a change in failure patterns, with isolated LF being more common in the 3D-CRT group and isolated out-of-field failures predominating in the tomotherapy group.

Geometric validation of MV topograms for patient localization on TomoTherapy.

Our goal was to geometrically validate the use of mega-voltage orthogonal scout images (MV topograms) as a fast and low-dose alternative to mega-voltage computed tomography (MVCT) for daily patient localization on the TomoTherapy system. To achieve this, anthropomorphic head and pelvis phantoms were imaged on a 16-slice kilo-voltage computed tomography (kVCT) scanner to synthesize kilo-voltage digitally reconstructed topograms (kV-DRT) in the Tomotherapy detector geometry. MV topograms were generated for couch speeds of 1-4 cm s(-1) in 1 cm s(-1) increments with static gantry angles in the anterior-posterior and left-lateral directions. Phantoms were rigidly translated in the anterior-posterior (AP), superior-inferior (SI), and lateral (LAT) directions to simulate potential setup errors. Image quality improvement was demonstrated by estimating the noise level in the unenhanced and enhanced MV topograms using a principle component analysis-based noise level estimation algorithm. Average noise levels for the head phantom were reduced by 2.53 HU (AP) and 0.18 HU (LAT). The pelvis phantom exhibited average noise level reduction of 1.98 HU (AP) and 0.48 HU (LAT). Mattes Mutual Information rigid registration was used to register enhanced MV topograms with corresponding kV-DRT. Registration results were compared to the known rigid displacements, which assessed the MV topogram localizations sensitivity to daily positioning errors. Reduced noise levels in the MV topograms enhanced the registration results so that registration errors were <1 mm. The unenhanced head MV topograms had discrepancies < 2.1 mm and the pelvis topograms had discrepancies < 2.7 mm. Result were found to be consistent regardless of couch speed. In total, 64.7% of the head phantom MV topograms and 60.0% of the pelvis phantom MV topograms exactly measured the phantom offsets. These consistencies demonstrated the potential for daily patient positioning using MV topogram pairs in the context bony-anatomy based procedures such as total marrow irradiation, total body irradiation, and cranial spinal irradiation.

SU-E-T-177: Experience Based Predicition Model For Automated VMAT Planning: A Cervical Cancer Application

PURPOSE To develop a planning prediction model based on patient geometry and dose-volume endpoins from previously treated cases; to devise an automated volumetric modulated arc therapy (VMAT) planning for future cases. METHODS Thirty cervical patients were included for this retrospective study, including22 patients fortraining cohort and 8 for testing cohort. For each patient in the training cohort, a VMAT plan with two full arcs weregenerated using a clinical plan template. The relative volume of the selected organ-at-risks (OARs) within a specified margin of the PTV(Lx) were extracted from overlap volume histograms (OVHs). A prediction model at 2D dose-distance (Lx, Dx) grid were established using a linear regression model. For the testing cohort,the model predicted DVH endpoints were used as constraints to automatically generate a new VMAT plan, the new plans were evaluated against the original plans. RESULTS On average, the prediction doses of rectum, bladder and bowel were 1.13Gy, 2.09Gy, 0.81Gy lower than the manual VMAT plans at predicted DVH endpoints. The auto plan showed slightly lower PTV conformity,CI = 1.18(auto) vs CI =1.12(manual) and almost the same PTV uniformity (UI= 0.17(auto) vs UI=0.18(manual)). V40 of rectum and bladder and V30 of bowelfrom auto plan were reduced by 2.36%, 13.56% and 4.36%, respectively. Mean doses of the rectum, bladder and bowel reduced 0.35Gy, 2.17Gy and 1.08Gy, respectively, as compard with the manual plans. CONCLUSION The experience-based prediction model has demonstrated the ability as a plan quality control tool to further improve OARs dose constrains setting in optimization process, butoffers a potential method for generating automatic VMAT plans which significantly improve theplanqualityand planning efficiency.

SU‐D‐12A‐02: DeTECT, a Method to Enhance Soft Tissue Contrast From Mega Voltage CT

PURPOSE MVCT images have been used on TomoTherapy system to align patients based on bony anatomies but its usefulness for soft tissue registration, delineation and adaptive radiation therapy is severely limited due to minimal photoelectric interaction and prominent presence of noise resulting from low detector quantum efficiency of megavoltage x-rays. We aim to utilize a non-local means denoising method and texture analysis to recover the soft tissue information for MVCT. METHODS A block matching 3D (BM3D) algorithm was adapted to reduce the noise while keeping the texture information of the MVCT images. BM3D is an imaging denoising algorithm developed from non-local means methods. BM3D additionally creates 3D groups by stacking 2D patches by the order of similarity. 3D denoising operation is then performed. The resultant 3D group is inversely transformed back to 2D images. In this study, BM3D was applied to MVCT images of a CT quality phantom, a head and neck and a prostate patient. Following denoising, imaging texture was enhanced to create the denoised and texture enhanced CT (DeTECT). RESULTS The original MVCT images show prevalent noise and poor soft tissue contrast. By applying BM3D denoising and texture enhancement, all MVCT images show remarkable improvements. For the phantom, the contrast to noise ratio for the low contrast plug was improved from 2.2 to 13.1 without compromising line pair conspicuity. For the head and neck patient, the lymph nodes and vein in the carotid space inconspicuous in the original MVCT image becomes highly visible in DeTECT. For the prostate patient, the boundary between the bladder and the prostate in the original MVCT is successfully recovered. Both results are visually validated by kVCT images of the corresponding patients. CONCLUSION DeTECT showed the promise to drastically improve the soft tissue contrast of MVCT for image guided radiotherapy and adaptive radiotherapy.

International Multi-Institutional Bench Mark Study on Dosimetric and Volumetric Modulation Using Helical TomoTherapy Treatment Planning for Malignant Pleural Mesothelioma Tumors

determining the most desirable and achievable target dose and organ at risk (OAR) sparing using helical TomoTherapy planning system for mesothelioma treatment plans. A range of planning parameters was used. The reviewers’ ranking assessment (Ranking in Groups: 1 = Good, 2 = Above Average, 3 = Average, 4 = Poor).The overall rankings revealed that a plan with a balanced tradeoff among all planning objectives was preferred by most participants and reviewers. Other studies found low doses to the contralateral lung to be limiting. This was not the case in our study, with TomoTherapy we found the dose to contra lateral lung to be as low as V5Gy=0.87%. A pitch value of 0.287 or 0.43 would provide better result. A delivered modulation factor of above 1.7 and a treatment time around 500 sec will be beneficial consideration in planning.

SU-E-T-532: Left-Sided Breast Cancer Irradiation Using Volumatric Modulated Arc Therapy: An Evaluation of Multiple Commercial Systems

Purposes: There has been growing interest in treating breast cancer using VMAT technique. Our goal is to compare the dosimetry and treatment delivery parameters for the left-sided breast cancer treatment using various VMAT platforms from commercially available planning systems. Methods: Five consecutive left-sided breast cancer patients initially treated with conventional 3D-conformal radiotherapy (3DCRT) were selected. Four VMAT plans using most popular treatment planning systems, including Eclipse (Version 11, Varian), Pinnacle (Version 9.8, Philips), Monaco (Version 2.03, Elekta) and helical Tomotherapy (V4.0, Accuray). The same structure set and same planning goals were used for all VMAT plans. The dosimetric parameters including target coverage and minimum/maximum/mean, dose-volume endpoints for the selected normal structures: the heart, ipsilateral-/contralateral lung and breast, were evaluated. Other dosimetric indices including heterogeneity index (HI) were evaluated. The treatment delivery parameters, such as monitor unit (MUs) and delivery time were also compared. Results: VMAT increases dose homogeneity to the treated volume and reduces the irradiated heart and left-lung volumes. Compared to the 3DCRT technique, all VMAT plans offer better heart and left-lung dose sparing; the mean heart doses were 4.5±1.6(Monaco), 1.2±0.4(Pinnacle), 1.3± (Eclipse) and 5.6±4.4(Tomo), the mean left-lung doses were 5.9±1.5(Monaco), 3.7±0.7(Pinnacle), 1.4± (Eclipse) and 5.2±1.6 (Tomo), while for the 3DCRT plan, the mean heart and left-Lung doses were 2.9±2.0, and 6.8±4.4 (Gy) respectively. The averaged contralateral-breast and lung mean doses were higher in VMAT plans than the 3DCRT plans but were not statistically significant. Among all the VMAT plans, the Pinnacle plans often yield the lowest right-lung/breast mean doses, and slightly better heterogeneity indices that are similar to Tomotherapy plans. Treatment delivery time of the VMAT plans (except helical Tomotherapy IMRT) is estimated to be comparable with the conventional 3DCRT. Conclusion: VMAT achieves equal or better PTV coverage and comparable OARs sparing compared to the conventional 3DCRT techniques.

SU-F-BRB-13: Correlation of Improved Target and Organ-At-Risk Dosimetric Quantities and Clinical Outcomes for Helical Tomotherapy Treated Mesothelioma

Purpose: We have observed improved local control probability (LCP) for adjuvant mesothelioma radiotherapy following pleurectomy/decortication using Tomotherapy compared to the conventional 3D technique (p<0.05). This work assesses the correlation between the improved clinical outcomes against dosimetry quantities. Methods: Thirty-eight mesothelioma cases consecutively treated at our clinic were retrospectively analyzed. Sixteen patients were treated using 3D technique planned on the Eclipse for c-arm accelerators prior to 7/2012; the other 22 cases were treated on Tomotherapy using helical IMRT after 7/2012. Typical 3D plans consisting of 15 MV AP/PA photon fields prescribed to 10 cm depth followed by matching electron fields with energy ranging from 8–16 MeV. Tomotherapy plans were designed using 2.5cm jaw, 0.287 pitch with directional blocking of the contralateral lung. The same prescription of 45 Gy (1.8GyX25) was used for both techniques. The dosimetry metrics for the critical structures: ipsilateral-/contralateral-lung, heart, cord, esophagus, etc were compared between two techniques. Results: Superior LCP is closely associated with improved target coverage. Tomotherapy plans yielded dramatically better target coverage and less dose heterogeneity despite of more advanced/larger disease. The averaged PTV volumes were 2287.3±569.9 (Tomotherapy) vs. 1904.8±312.3cc (3D); V100s were: 91.1±4.0 (%) vs. 47.8±12.7 (%) with heterogeneity indices of 1.20±0.1 vs.1.37±0.38 and for the Tomotherapy and 3D plans, respectively. Compared to the 3D technique, we observed significant lower maximum cord doses (p<0.001), lower mean esophagus doses (p<0.002), and lower heart mean doses when tumor was left-sided (p=0.002). For ipsilateral-/contralateral-lungs, however, the mean doses and V20, V5 of Tomotherapy plans were significantly higher than the 3D plans (p<0.01) regardless which sides of lung were treated. However, rates of radiation pneumonitis were no different. Conclusion: Tomotherapy achieved great improvement of plan quality including target coverage, resulting in significantly better local control over the traditional 3D technique for adjuvant radiotherapy for mesothelioma.

SU‐E‐T‐01: (In)dependence of Plan Quality On Treatment Modalities and Target‐To‐Critical Structure Geometry for Brain Tumor

Purpose: To evaluate and test the hypothesis that plan quality may be systematically affected by treatment delivery techniques and target-tocritical structure geometric relationship in radiotherapy for brain tumor. Methods: Thirty-four consecutive brain tumor patients treated between 2011–2014 were analyzed. Among this cohort, 10 were planned with 3DCRT, 11 with RadipArc, and 13 with helical IMRT on TomoTherapy. The selected dosimetric endpoints (i.e., PTV V100, maximum brainstem/chiasm/ optic nerve doses) were considered as a vector in a highdimensional space. A Pareto analysis was performed to identify the subset of Pareto-efficient plans.The geometric relationships, specifically the overlapping volume and centroid-of-mass distance between each critical structure to the PTV were extracted as potential geometric features. The classification-tree analyses were repeated using these geometric features with and without the treatment modality as an additional categorical predictor. In both scenarios, the dominant features to prognosticate the Pareto membership were identified and the tree structures to provide optimal inference were recorded. The classification performance was further analyzed to determine the role of treatment modality in affecting plan quality. Results: Seven Pareto-efficient plans were identified based on dosimetric endpoints (3 from 3DCRT, 3 from RapicArc, 1 from Tomo), which implies that the evaluated treatment modality may have a minor influence on plan quality. Classification trees with/without the treatment modality as a predictor both achieved accuracy of 88.2%: with 100% sensitivity and 87.1% specificity for the former, and 66.7% sensitivity and 96.0% specificity for the latter. The coincidence of accuracy from both analyses further indicates no-to-weak dependence of plan quality on treatment modality. Both analyses have identified the brainstem to PTV distance as the primary predictive feature for Pareto-efficiency. Conclusion: Pareto evaluation and classification-tree analyses have indicated that plan quality depends strongly on geometry for brain tumor, specifically PTV-tobrain-stem-distance but minimally on treatment modality.

WE-E-BRE-10: Level of Breast Cancer Stem Cell Correlated with Tumor Radioresistence: An Indication for Individualized Breast Cancer Therapy Adapted to Cancer Stem Cell Fractions

PURPOSES The presence of cancer stem cells (CSCs) in a solid tumor could result in poor tumor control probability. The purposes are to study CSC radiosensitivity parameters α and β and their correlation to CSC levels to understand the underlying radioresistance mechanisms and enable individualized treatment design. METHODS Four established breast cancer cell lines (MCF-7, T47D, MDA-MB-231, and SUM159PT) were irradiated in vitro using single radiation doses of 0, 2, 4, 6, 8 or 10 Gy. The fractions of CSCs in each cell lines were determined using cancer stem cell markers. Mammosphere assays were also performed to better estimate the number of CSCs and represent the CSC repopulation in a human solid tumor. The measured cell surviving fractions were fitted using the Linear-quadratic (LQ) model with independent fitting parameters: α_TC, β_TC (TCs), α_CSC, β_CSC (CSCs), and fs (the percentage of CSCs in each sample). RESULTS The measured fs increased following the irradiation by MCF-7 (0.1%), T47D (0.9%), MDA-MB-231 (1.18%) and SUM159T (2.46%), while decreasing surviving curve slopes were observed, indicating greater radioresistance, in the opposite order. The fitting yielded the radiosensitive parameters for the MCF-7: α_TC=0.1±0.2Gy-1 , β_TC= 0.08 ±0.14Gy-2 , α_CSC=0.04±0.07Gy-1 , β_CSC =0.02±0.3Gy-2 ; for the SUM159PT, α_TC=0.08±0.25 Gy-1 , β_TC=0.02±0.02Gy-2 , α_CSC=0.04±0.18Gy-1 , β_CSC =0.004±0.24Gy-2 . In the mammosphere assay, where fs were higher than the corresponding cell line assays, there was almost no shoulder found in the surviving curves (more radioresistant in mammosphere assays) yielding β_CSC of approximately 0. CONCLUSION Breast cancer stem cells were more radioresistant characterized by smaller α and β values compared to differentiated breast cancer cells. Percentage of breast cancer stem cells strongly correlated to overall tumor radioresistance. This observation suggested the feasibility of individualized radiotherapy prescription based on the fractions of cancer stem cells found in biopsy.