Planning CT Research Articles

Quantitative Cone-Beam CT Imaging in Radiotherapy: Parallel Computation and Comprehensive Evaluation on the TrueBeam System

Cone-beam CT (CBCT) imaging is used in the patient setup on the advance image-guided radiation therapy. However, the scatter contamination causes spatial non-uniformity, the error of the CT number, and image contrast loss, which is considered as one of the fundamental limitations for CBCT application. In this paper, we use both scatter correction and noise suppression for improving the quality of CBCT image and comprehensively evaluate our method on patient data. A CBCT software package with parallel computation is designed, which is easily compatible with the existing CBCT system of radiotherapy device with high computing efficiency and CT number accuracy. The primary signals of projections are estimated through the use of the forward projections from the registered planning CT (pCT). The errors of low frequency in the raw projections are obtained through subtracting the forward projections and the low-pass filter implementation. The penalty-weighted-least-square (PWLS) method is applied to reduce the high-frequency noise in the corrected CBCT projections. We use the graphics-processing unit (GPU) NVidia Tesla C2075 card with CUDA C programming to accelerate the time-consuming processes. The CBCT projections of a pelvis phantom and the two pelvis patients are obtained from the Varian TrueBeam system, which is a machine for the radiotherapy. The maximum errors of CT number are reduced over 70 HU on the TrueBeam result to below 15 HU, and the errors of spatial non-uniformity are decreased by a factor of around 7. The computation time is about 25 min on the GPU, which is reduced over 10 h on the CPU. The proposed software shows superior performance over the existing reconstruction. The proposed software demonstrates the reliability of the high-accuracy CBCT-based image-guided radiotherapy (IGRT), providing the high-precise CBCT with less computation time.

27 CBCT-based dose calculation for head-and-neck tumors treated with VMAT in order to approach adaptive radiotherapy

Introduction Head-and-neck (H&N) treatment delivery time is divided in 35 fractions of 2 Gy and takes 7 weeks. Anatomical modifications can occur during treatment and introduce uncertainties in VMAT dose distribution. Cone Beam Computed Tomography (CBCT) is used for patients positioning but might be considered for dose calculation. The purpose of this study is to improve the image quality and accuracy of CBCT-based dose calculation through the use of an optimized CBCT acquisition mode to finally build a patient-specific H&N deformation model to approach adaptive radiotherapy (ART). Methods A calibration was performed with the view to achieve a higher accuracy in a 25 cm field-of-view (FOV) CBCT ORL , ART (Varian Clinac® OBI 1.6) image-based dose assessment by varying the source-to-imager distance (SID), tube voltage (kVp) and mAs. Then, the Hounsfield unit (HU) vs relative electron density curves were drawn for the CIRS (062 M) phantom associated to the QuasarTM (Modus QA) using the post-treatment software Syngo.via (Siemens Healthinners) with the intention of comparing them to the planning CT curve ( CT plan ) . The optimized CBCT ORL , ART acquisition mode dose has been assessed as described in the AAPM report n. 111 [1] with the use of a 0.3 cm3 ionization chamber inside a 26 cm PMMA cylindrical phantom. This retrospective cohort study includes 38 H&N cancer patient cases who underwent VMAT treatment with one CBCT ORL , ART per week in 2017. For each patient, a deformable image registration algorithm (DIR) has been applied on the CT plan and on the 7 CBCT ORL , ART using Syngo.via. A comparative study between deformation vector fields (DVF), planned target volume (PTV) and organ-at-risk (OAR) contours similarity measurements and dose-volume histogram (DVH) has been done with the aim of assessing the feasibility of CBCT-based dose calculation. Results The optimized CBCT ORL , ART mode chosen had the following acquisition parameters: 100 kVp, 145 mAs and a 150 cm SID. A 3.83 mGy CTDIw has been assessed for the CBCT ORL , ART mode with a + 1.8% relative gap between measured and constructor data. Due to the limited field of view (FOV), truncation of the CBCT caused incorrect deformations when the DIR algorithm was applied and led to significant errors in dose calculation while deforming the CT plan image to match the CBCT ORL , ART was considered suitable for dose calculation based on the 2 mm gamma-test on over 95% of the voxels and DVH. Conclusions This first adaptive radiotherapy approach verifies the feasibility of CBCT-based dose calculation. Following this study, a dosimetric alert protocol was set up to evaluate the need of performing a re-plan during the treatment.

178. Trial prospettico per predire la correlazione dose risposta nei muscoli costrittori faringei in pazienti con tumore testa-collo Prospective trial to predict the pharyngeal constrictor muscle dose-response correlation in head and neck cancer patients

PurposeRadiotherapy, alone or combined with chemotherapy, is a treatment modality used frequently in head and neck cancer. It is important to delineate organs in the head and neck at risk for radiation-induced swallowing dysfunction (RISD). Irradiation of the Pharyngeal Constrictor Muscle (PSCM) seems to play a crucial role in swallowing dysfunctions and in particular in the voice and speech quality changes. Purpose of our retrospective study was to identify which dose volume histogram parameters are most important to predict patient radiation-induced swallowing dysfunction, voice and speech quality. MethodsPatients with head and neck cancer who previously underwent Tomotherapy treatment are selected. The prescribed radiotherapy is based on SIB technique for all PTVs irradiated simultaneously over 30 daily. Doses of 67.5, 60 and 54–57 Gy were prescribed to primary tumour, high-risk nodal regions and low-risk nodal regions, respectively. Upper PSCM, middle PSCM and lower PSCM were outlined separately and DVHs are stored for each patient using basal planning CT and weekly MVCT. Univariate and multivariate analysis will be used to identify novel dosimetric predictors of toxicity and predictors of quality of life assessed using appropriate questionnaires taking into consideration the actual delivered dose. ResultsThe study started in July 2017 and the up to now 8 patients have been included in the study. Preliminary results suggest the importance of taking into consideration the dosimetric information based on basal planning CT and MVCT in order to improve the evaluation of toxicity rate and occurrence, which are still ongoing. ConclusionsDosimetric factors based on planning CT and MVCT seem to improve the identification of appropriate predictors of toxicity. More patients are to be included in the study to fully confirm these preliminary results.

A prospective study of weekly intensity modulated radiation therapy plan adaptation for head and neck cancer: improved target coverage and organ at risk sparing.

This prospective study of weekly CT scanning and plan adaption during H&N IMRT reports on the frequency of plan adaptations based on dosimetric differences between original and re-optimised IMRT plans. The volumetric and geometric change occurring in target volumes and salivary glands is also described. Ten H&N cancer patients underwent weekly planning CT imaging and re-optimisation of the IMRT plan if PTV or OAR coverage was unacceptable. Comparisons of PTV and parotid gland dosimetry between the original and adaptive plans were made. Parotid and submandibular gland volume changes and shift were calculated. Eight of ten patients required one or more plan adaptations, with 41% of adaptations occurring by fraction ten. Salivary glands reduced in volume, with a medial shift of the lateral border of the parotid gland and a superior shift of the submandibular gland. Change in PTV coverage did not correlate with weight loss or nutritional score. Inadequate PTV coverage, requiring plan adaptation, occurs early in the course of IMRT. A weekly Adaptive RT (ART) protocol results in significant improvement of PTV coverage. Implementation of a clinical ART protocol should include imaging and dose calculation within the first ten fractions.

Feasibility of MR-Generated Synthetic CT Images for Planning Stereotactic Radiosurgery of Brain Metastases

Frameless Linac-based stereotactic radiosurgery (SRS) treatment planning typically involves spatial registration of MR and CT images. A solely MR-guided workflow would eliminate the burdens of CT scanning and associated time as well as alignment errors, but would require dose calculations and patient positioning based on MR-derived information. The feasibility of MR-guided SRS of brain metastases was investigated. Specifically, it was hypothesized that synthetic CT images (MRCTs) could be used to generate plans and reference images for positioning that would not differ from SRS using a traditional CT-guided technique. MR and CT images were acquired for 19 patients undergoing SRS for brain metastases. Following rigid image alignment, PTV and OAR contours were delineated. MRCT image volumes were generated using a voxel-based probabilistic classification technique and aligned rigidly to their respective CT scans. The CT structures were transferred onto the MRCT volumes. CT-based plans were created with the intent to deliver 99.5% of the prescription dose to the PTV and avoid excess doses to the OARs. The fluences from the CT plans then were used for dose calculations on a “maskless” CT plan (in which the attenuation of the immobilization device was removed) and MRCT images. PTV D99.5 and OAR D0.1 cc values were compared. OARs (brainstem, optic nerves/chiasm, eyes, lenses, cochleae) typically defined on CT were drawn on the source MR (T1- and T2-weighted image volumes) images and compared to the original CT volumes. Cone beam CT (CBCT) volumes used for positioning at treatment were aligned to the MRCT and CT volumes, and resulting transforms compared. A total of 43 PTVs were studied with a mean prescription dose of 21.1 Gy (range 12 to 30 Gy). The mean difference in PTV D99.5 between the CT versus maskless CT was 0.0 % (range -0.0 to 0.1 %) and CT versus MRCT was 0.0 % (range -0.1 to 0.0 %) of the prescription doses. Dose metrics for a total of 185 OARs were calculated. OAR D0.1 cc difference between the CT versus maskless CT was 0.0 Gy (range -0.1 to 0.0 Gy) and CT versus MRCT was 0.0 Gy (range -0.2 to 0.7 Gy). For OAR volumes larger than 1 cc (n=64), the mean Dice coefficient between CT and MR structures was 0.82 (range 0.17 to 0.93). Mean differences in alignment between the CBCT volumes and MRCT and CT volumes resulted in isocenter positioning differences of 0.0 mm (range -0.2 to 0.1 mm), 0.0 mm (range -0.5 to 0.4 mm), and 0.1 mm (range -0.2 to 0.5 mm) in the LR, AP, and CC directions. The accuracy of delineation, calculated doses, and patient positioning indicate that MR-guided frameless SRS is feasible. Future investigations will explore generation of reference image volumes for alignment that combine MR and synthesized CT contrasts. Supported by NIH R01EB016079.

Geometric Reproducibility of Fiducial Markers and Efficacy of a Patient-Specific Margin Design Using Active Breath Hold for Pancreas Stereotactic Body Radiation Therapy

In patients undergoing SBRT for pancreatic adenocarcinoma using an active breathing control (ABC), the reproducibility of fiducial positioning between breath-holds is unclear. We characterized this variation by acquiring multiple breath-hold CTs at time of simulation and analyzed whether a patient-specific margin based on this variation would help account for intra-fraction variation at time of treatment. We analyzed five pancreatic cancer patients who underwent SBRT for pancreatic adenocarcinoma and who were treated with ABC. At time of simulation, 3-additional breath-hold scans were acquired immediately after the primary contrast-enhanced planning CT with 2-mm slice thickness. To characterize the variability in tumor position between breath-holds, fiducial centroids from the planning CT were used as the reference and compared to the fiducial centroids from the additional breath-hold CTs. Both the average excursions in fiducial position (Sim Varavg) and the maximum excursions in fiducial position (Sim Varmax) were calculated. For treatment planning for each patient, the amount of variation in each direction between breath-holds was applied as a margin to the GTV, followed by an additional 2-mm margin to determine the final PTV. The difference in volume between this method versus directly expanding the GTV by a uniform 2-mm expansion was calculated. During 5-fraction treatment, 20-27 breath-hold CBCTs were acquired for each patient. The intrafraction variation in fiducial position was calculated using the same method as above, using the initial CBCT for that fraction as the reference CT and using similar metrics to characterize fiducial position variation (Tx Varavg and Tx Varmax). The adequacy of a direct 2-mm expansion on the GTV was analyzed by evaluating the percentage of fiducial locations within 2-mm of variability on CBCTs. Table 1 summarizes the breath-hold reproducibility averaged over all five patients at CT sim and treatment. By applying the variation seen during multiple breath-holds at time of simulation to the GTV volume before further expanding by 2 mm, the final PTV volume increased by 19.5 ± 7.0% as compared to a uniform 2-mm expansion off of the GTV. However, by using this patient-specific margin, 95% of the all fiducial locations during treatment were encompassed within this volume, while a uniform 2-mm expansion only covered the fiducials 80% in both LR and AP directions and 60% in the SI direction.Abstract SU_18_2175; Table 1Inter-BH reproducibilityLR (mm)AP (mm)SI (mm)Sim Varavg0.8 ± 0.51.0 ± 0.41.5 ± 0.7Sim Varmax1.2 ± 0.71.4 ± 0.52.3 ± 0.6Tx Varavg1.2 ± 0.51.4 ± 0.62.1 ± 0.4Tx Varmax4.7 ± 2.15.5 ± 3.06.4 ± 2.4 Open table in a new tab Variation in tumor position exists between breath-holds when treating with ABC. Acquiring multiple breath-hold CTs during simulation can help quantify inter-breath-hold reproducibility and can provide a patient specific margin design to compensate for the variation in tumor position between breath-holds at time of treatment.

Analysis of the Dosimetric and Geometrical Variations Caused by Irregular Breathing Patterns on Normal Helical CT and 4D CT Datasets for Lung Cancer SBRT Treatment

To analyze the dosimetric and geometrical variations due to irregular breathing patterns on normal helical CT and 4D CT datasets for lung cancer SBRT treatments. Regular and irregular motion models were created and imported into a dynamic thorax phantom for imaging, planning and does measurement 64-slice CT, 6-MV photon beams and a treatment planning system were used. The regular motion was a sine wave with longitudinal range of ±1.5 cm and period of 4 sec. The irregular motions were generated by extracting the RPM cycles of different degrees of amplitude or phase variations from patients. Two 20 mm diameter and soft-tissue equivalent semispherical target was embedded in a motion rod for use with the phantom. Helical CT for static target, and 4D CT for motion target in regular and all irregular motions were performed. Phase bin, maximum and average intensity projection (MIP, AIP) CTs were reconstructed to evaluate the target position and geometric distortion. Static plan, and the motion plans for different patterns with phase-based gating (interval of 30–70%) and non-gating treatment were generated using RaipArc and IMRT techniques. EBT3 film was used for dose measurement. Dose profile comparison and gamma analysis (±3%/1 mm criteria) were used to quantify the dose agreement between measurements and calculations. Compare with static CT, MIP images shown a slight higher HU (value of 50) in the target region. By contrast to MIP images with a homogeneous HU distribution in the target region, AIP images were with a much lower HU and with a shape of Gauss distribution (-780 to -350). Phase bin with 10 phases or 20 phases have similar geometrical shapes. Phase bin images with higher target motion velocity (phase 25% and 75%) have a significantly larger target center position errors (3.0 to 3.4 mm). The variations of target volume for static CT and regular motion phase bin CTs were within 2.5%, AIP and MIP CTs showed a volume reduction of -6.7% and -3.8%, respectively. For irregular motions, slightly differences (-6.2 to -7.7%) on ITV between AIP and MIP images (AIP/MIP) were found for gated plans, and had a larger differences (-12.3 to -15.2%) for non-gated plans. Dosimetric comparisons showed a high passing rate (>98.5%) for static plan in the target region, AIP and MIP gated plans have average passing rates of 92.2±5.3% and 85.8±8.9%, respectively. Non-gated plans have significantly lower and deviated passing rates, AIP and MIP plans were with passing rates of 43.6±20.8% and 66.7±36.4%, respectively (p < 0.05). Faster motion results in larger position errors in 4D CT image acquisition, and Irregular breathing motion leads to volume reduction for AIP images; therefore, plan CT should avoid using fast period CT, breath coaching should be used to maintain regular motion, and the ITV in AIP CT should be verified. For Irregular motions during treatment, phase-based gated plans have target doses more consistent with the plans than the non-gated plans.

On-Board Cone Beam Computed Tomography as Radiation Therapy Simulation for Emergent Cases

To investigate dosimetric clinical appropriateness of on-board kV cone beam computed tomography patient simulation (CBCT-Sim) for palliative treatment plans at a remote satellite clinic having no conventional computed tomographic simulation (CT-Sim) when travel isn't a viable option. Our current approach is a separation calculation using two opposing fields on a water phantom. A CBCT of the patient would improve the plan by utilizing the patient's actual geometry and we expect to see dose points in the CBCT-Sim plan image set less than 3% different than a conventional CT, and the separation calculation to be within 5% of the conventional CT-Sim plan. Patients treated for the four most likely emergent or immobile disease sites (brain, thoracic spine, lumbar spine, and pelvis) with an original CT-Sim and at least one CBCT during course of treatment were investigated retrospectively. Three brain patients, three T-spine, three L-spine, and two pelvis patients were used. The patients’ images were copied into a new, anonymous patient. A new CT-density curve was created in the treatment planning software for the CBCT based on CT-density phantom data. A standard plan for each given site was planned on the original CT image set, then copied on both a heterogenous and homogenous CBCT image set. The homogenous CBCT image was density overridden to water. A rectangular water phantom was also created for a separation calculation plan, which is our standard practice for these cases. For the brain cases, the CT and heterogenous CBCT plans were both contoured with critical structures. For all cases, seven to eleven points of interest in all directions (isocenter, and fractional separation distances in 3-dimentional space) were used to compare both CBCT-Sim plans and the separation calculation to the standard CT plan. For all cases, the average dose difference per case compared to the standard CT-Sim plan was less than 3% for either CBCT-Sim plan and less than 4% for the separation calculation plan; however, the separation calculations showed large deviations, up to 7.2%, while the CBCTs showed a maximum of 4.8%. Between all cases, the average dose difference was lowest compared to CT-Sim plans for the heterogenous CBCT plan with 1.2±0.7%, while the homogenous CBCT plan difference was 1.4±0.7% and the separation calculation was 1.7±1.1%. The most dosimetric benefit for the heterogenous CBCT-Sim plans was seen for the brain plans, followed by the T-spine plans, with negligible benefit for pelvis and L-spine plans. Dosimetric accuracy is improved with a CBCT-Sim compared to our standard separation-based plan, due to the use of actual patient geometry and density.

Efficacy of Epid-Based In Vivo Dosimetry and Calibrated CBCT Images for a Timely Lung Cancer Replanning

The tumor volume shrinkage during the treatment of lung cancer can require an adaptive plan. The aim of this study is to investigate the efficacy of electronic portal imaging device (Epid) in vivo dosimetry (IVD) combined with calibrated cone beam computed tomography (CBCT) scans for a timely replanning. Sixteen lung cancer patients treated with volumetric modulated arc therapy were enrolled for this study. They underwent CBCT and Epid-based IVD verification twice a week and a second scan (CT2) in the middle of the treatment for a morphological reevaluation. The accuracy and reproducibility of the treatment delivered was assessed with: i) IVD tests by a quality control program that supplies three indices: the ratio R between the reconstructed and planned isocenter doses and g%, gmean (gamma analysis of the electronic portal images). They were considered in tolerance if R within 5%, g % >90% and if gmean <0.4; ii) a recalculation of the reference treatment planning (TPref) using CBCT scans acquired along the treatment and calibrated by the same quality control software. Therefore, the agreement of the DVH dosimetric indices D95%(PTV), V20Gy (lung), D1cc (spinal cord), Dmean(heart) with the ones computed on the planning CT(CTplan) was verified; a DVH indices difference of 3% was considered acceptable. IVD tests resulted out of tolerance for three patients before the scheduled CT2; DVH indices obtained by the plan recalculation on calibrated CBCT confirmed this discrepancy. For these patients, TPa was carried out on CT2 and compared with TPa computed on CBCT scans. In the table below, the results obtained for one of these three patients treated with 60Gy in 20 fractions over four weeks for a lung metastasis were reported; IVD tests and DVH indices out of the tolerance level were highlighted in bold. The warnings supplied by Epid-based IVD can offer an early advice for a replanning step. The adaptive plan computed on calibrated CBCT can be timely performed and can provide DVH indices with an accuracy of 3%.Abstract TU_29_3183; Table 1IVD Test ResultsDose CBCT vs CT (DVH indices difference)weekCT scan/ Treatment PlanRg%gmeanPTV (D95%)Homolateral Lung (V20Gy)Spinal Cord (D1cc)Heart (Dmean)week 1CTplan/TPref0.9795%0.171.1-1.71.21.9week 1CTplan/TPref0.96100%0.020.8-2.51.21.9week 2CTplan/TPref0.9998%0.093.12.31.21.9week 2CTplan/TPref0.9481%0.417.06.6-0.92.3week 3CTplan/TPref0.9386%0.358.05.90.60.9week 3CT2/TPa1.0098%0.082.12.11.0-2.2week 4CT2/TPa0.9996%0.091.00.90.0-2.8week 4CT2/TPa0.9996%0.050.81.8-1.1-2.0 Open table in a new tab

Deformation-Guided Hybrid Approach for Interfractional Variations

It has been well documented that interfraction variations (primarily organ deformation) that cannot be fully accounted for by image-guided radiation therapy (IGRT) (i.e., patient repositioning) but can be corrected using online adaptive replanning (OLAR). However, OLAR is time consuming and labor intensive compared to the repositioning process. A hybrid approach that can uses either OLAR or repositioning would be practical as OLAR is not necessary for all fractions. The major issue is to rapidly and objectively determine if OLAR is needed after daily image is acquired. To address issue, we investigate whether the magnitude of deformation can be used to guide the selection of OLAR versus repositioning. Daily CTs acquired using an in-room CT during radiation therapy for 10 representative patients with prostate cancer were deformably registered with reference CTs. The deformable registration was verified in accordance with the AAPM TG-132 report. The deformation vector field (DVF) was quantified using the Jacobian determinant matrix. Plans were created with a prescribed dose of 36.25 Gy to 95% of the planning target volume (PTV) of prostate plus a 3 mm margin in five fractions. For each daily CT, repositioning and OLAR plans were generated. Five fractional doses from either a repositioning or an OLAR plan as indicated by the DVF were deformed back to the planning CT and accumulated for the hybrid approach. Doses from five repositioning plans (all repo) and five replanning (all OLAR) were also accumulated for comparison. The Jacobian determinants measured the magnitude of deformation between the reference and daily CTs. Appropriate threshold on the amount of deformation to indicate the necessity of OLAR was determined. Depending on the amount of deformation, a hybrid approach including OLAR and repositioning was used. Table I presents a comparison of the plan qualities of the hybrid approach, all repo and all OLAR. It can be seen that the hybrid approach as determined by the magnitude of deformation can generate plans comparable to those from all OLAR treatments. Analysis of interfraction anatomic deformation is a robust yet fast method to determine instances in which online adaptive replanning is necessary. This process can be completed without any segmentation, thus is fast (less than 1 minute), and allows us to quickly and objectively determine which fractions require adaptive replanning in a hybrid approach.Abstract TU_37_3258: Table IPTV V100 (%)OAR V50% dose (%)Hybrid (# repo)all OLARall repohybridall OLARall repoP1*-95.090.6-15.015.0P294.9 (1)95.094.716.716.119.5P395.0 (1)95.095.715.015.019.0P495.0 (1)95.087.222.722.721.6P595.2 (1)95.292.323.823.822.2P694.8 (2)95.790.38.18.19.3P795.0 (3)95.095.15.15.15.7P896.2 (3)96.294.423.923.923.2P995.5 (4)95.095.522.321.123.1P1095.0 (4)95.193.234.732.834.7*OLAR required for all fractions Open table in a new tab

Dose Mapping Local Failure Following Radical Image Guided Radiation Therapy to the Bladder and Pelvic Lymph Nodes

Most recurrences after radical bladder radiotherapy occur within the bladder. It is hypothesized that local failure occurs because of geographical miss, and or insufficient dose. Accurate mapping of recurrences in relation to the radiation dose is important in assessing the success of recent image guided bladder radiotherapy strategies and margin reduction. We aim to determine the dose prescribed and received to the volume that harbored the local failure. Patterns of local relapse were evaluated in 38 patients with high risk T2 (small cell cancer) or T3-4N0-3M0 muscle invasive bladder cancer recruited prospectively to an ethics approved phase II radiotherapy protocol treating whole bladder and elective pelvic lymph nodes to 52Gy, radiologically involved lymph nodes to 60Gy and bladder tumor to 64Gy in 32 fractions. Applied expansions are as outlined in Table 1. Soft tissue imaging was acquired prior to treatment with kV-CBCT. Use of neo-adjuvant and concurrent chemotherapy was recommended.Abstract SU_20_2199; Table 1Applied expansionsStructureApplied expansion to create corresponding PTV (cm)Planned dose (Gy)SuperiorInferiorLateralAnteriorPosteriorBladder tumor1.01.01.01.01.064Whole bladder1.50.50.51.51.052Involved nodes0.50.50.50.50.560Uninvolved nodes0.50.50.50.50.552 Open table in a new tab Time to local failure was defined from the start of radiotherapy to pathological confirmation of relapse, and was estimated using Kaplan Meier method. Cystoscopy and imaging (CT/MRI) were used to reconstruct the bladder relapse volume (GTVrelapse) on the initial planning CT. Dose to GTVrelapse (D98%) was determined from planning scan used for treatment. Twenty-seven of 38 (71%) of all patients had disease recurrence. Eighteen of 38 (47%) patients experienced first relapse within the bladder following radiotherapy. Median time to local relapse was 9.0 months (95% CI, 6.3-11.7 months). Fifteen of 18 (83.3%) of these patients received neo-adjuvant chemotherapy and 10/18 (56.6%) had received concurrent chemotherapy. Ten of 18 (55.6%) patients relapsed with ≥T2 local disease, 5/18 (27.8%) with pTa/pT1 disease, and 3/18 (16.7%) with carcinoma in situ. Fifteen of 18 (83%) of these relapses occurred within 1cm of the original bladder tumor site. Overall median GTVrelapse D98% was 57.1Gy (range 49.9-63.3Gy), 89.1% (77.9-99.0) of the prescribed dose (64Gy). Median GTVrelapse D98% where all or partial of the relapse volume was within 1cm of the original bladder tumor edge was 54.5Gy (range 51.4-63.3Gy), 85.2% (range 80.3-99.0%) of the prescribed dose (64Gy). Relapse patterns suggest local failure occurs within close proximity of the original bladder boost volume but outside the high dose treatment field (64Gy). Work is on-going to expand the patient numbers and to determine actual dose delivered to GTVrelapse D98% from on-line CBCT data. This will inform whether dose escalation, and or margin modification could improve reported outcomes.

Improving 3D-printing of megavoltage X-rays radiotherapy bolus with surface-scanner

BackgroundComputed tomography (CT) data used for patient radiotherapy planning can nowadays be used to create 3D-printed boluses. Nevertheless, this methodology requires a second CT scan and planning process when immobilization masks are used in order to fit the bolus under it for treatment.This study investigates the use of a high-grade surface-scanner to produce, prior to the planning CT scan, a 3D-printed bolus in order to increase the workflow efficiency, improve treatment quality and avoid extra radiation dose to the patient.MethodsThe scanner capabilities were tested on a phantom and on volunteers. A phantom was used to produce boluses in the orbital region either from CT data (resolution ≈1 mm), or from surface-scanner images (resolution 0.05 mm). Several 3D-printing techniques and materials were tested. To quantify which boluses fit best, they were placed on the phantom and scanned by CT. Hounsfield Unit (HU) profiles were traced perpendicular to the phantom’s surface. The minimum HU in the profiles was compared to the HU values for calibrated air-gaps. Boluses were then created from surface images of volunteers to verify the feasibility of surface-scanner use in-vivo.ResultsPhantom based tests showed a better fit of boluses modeled from surface-scanner than from CT data. Maximum bolus-to-skin air gaps were 1-2 mm using CT models and always < 0.6 mm using surface-scanner models. Tests on volunteers showed good and comfortable fit of boluses produced from surface-scanner images acquired in 0.6 to 7 min. Even in complex surface regions of the body such as ears and fingers, the high-resolution surface-scanner was able to acquire good models. A breast bolus model generated from images acquired in deep inspiration breath hold was also successful. None of the 3D-printed bolus using surface-scanner models required enlarging or shrinking of the initial model acquired in-vivo.ConclusionsRegardless of the material or printing technique, 3D-printed boluses created from high-resolution surface-scanner images proved to be superior in fitting compared to boluses created from CT data. Tests on volunteers were promising, indicating the possibility to improve overall radiotherapy treatments, primarily for megavoltage X-rays, using bolus modeled from a high-resolution surface-scanner even in regions of complex surface anatomy.

P3.16-18 Modern Radiotherapy Increases Patient Access to Curative Intent Radiotherapy in Non-Small Cell Lung Cancer

Technical developments in the delivery of radiotherapy such as image-guided radiotherapy (IGRT) and intensity modulated radiotherapy (IMRT) have permitted the introduction of advanced radiation techniques such stereotactic ablative radiotherapy (SABR). These new techniques have the advantage of more accurate localisation of the tumour and reduced irradiation of normal tissues. In our centre, we have implemented a range of new techniques to deliver IGRT (such as PET/CT and 4-dimensional CT planning, and cone beam CT during treatment delivery). We postulate that using the advanced techniques increases the access to curative intent radiotherapy treatment of lung cancer. We seek to assess the access rates to curative intent thoracic radiotherapy. Using our institutional lung radiotherapy database we analysed the data recording intent of treatment with reference to the stage and performance status (PS) of all patients with stage 1-3 non-small cell lung cancer (NSCLC) in 2007 and compared this to the same population receiving radiotherapy during 2017 and up to April 2018. In 2007, 217 patients with stages 1-3 NSCLC received any radiotherapy compared to 218 patients for the 2017/2018 cohort. Within the 2017/2018 cohort 96% of patients (n=94) received radical radiotherapy compared to 28% of patients in 2007 (n=26). Of the 94 patients receiving radical treatment in 2017/2018, 61% received SABR. This increase was largely due to in the introduction of SABR. In those patients with stage 3 disease, overall fewer patients received any radiotherapy in 2017/2018 compared to 2007, however the number of patients receiving curative intent radiotherapy increased from 20 (13% of all stage 3 patients) to 44 (37%). Of note in those patients receiving curative intent radiotherapy there was an increase in access for patients with poorer PS, with only 7% of patients with a PS ≥2 in 2007 receiving curative intent radiotherapy compared with 42% of patients in 2017/2018. In our series, we report an increase in the proportion of patients with NSCLC receiving curative intent radiotherapy. Furthermore, more patients with a poorer performance status received curative intent radiotherapy. We suggest that the introduction of advanced radiotherapy techniques has permitted the curative intent treatment of patients who were previously treated with a palliative approach to management.

Fluoroscopic Overlay of Landmarks from Pre-procedural CT Images after 2D3D Image Fusion to Optimize Trans-septal Access During Percutaneous Mitral Interventions

ABSTRACT Background: Recent evolution of transcatheter mitral devices and interventions has mandated the need for further optimizing trans-septal access technique with respect to its precise location and distance from the mitral valve annulus. Methods: Based on the individual procedural context, “ideal” trans-septal access site was planned in multi-slice CT images. In addition, relevant landmarks such as mitral annulus, para-valvular leak site, left atrial appendage (LAA) ostium were electronically marked. CT-fluoroscopy image fusion was performed using a commercially available software after acquiring two fluoroscopic images at-least 30° apart. Manual refinement of image fusion was performed by aligning structures such as a prosthetic valve or mitral annular calcification, when necessary. Results: Between June and October 2016, a total of five procedures (n = 3 Mitral paravalvular leak, n = 2 Mitraclip®) were performed in four patients using trans-esophageal echocardiography (TEE) and CT image fusion guidance. Fluoroscopic overlay of landmarks including the optimal trans-septal puncture site from pre-procedural CT images was feasible in all five procedures. Median time for CT-fluoroscopy image fusion was 4:08 minutes. Median time for left atrial access was 7:00 minutes. Median (range) fluoroscopy time and total procedural dose-area-product were 29.9 minutes (15.1–77.5) and 16,638 microGy-m2 (9464–18,990) respectively. Conclusion: In addition to real-time TEE imaging, pre-procedural CT planning, image fusion and overlay of relevant landmarks enhances 2D fluoroscopy and provided additional 3D image guidance. Seamless integration of such multi-modality image information can help further optimize steps such as trans-septal access for complex mitral interventions.

OA129] Calcifications in lung cancer patients: can they be used as surrogate for overall survival predictions?

Purpose Lung cancer patients treated with radiotherapy often present with multiple comorbidities which can interact with incidental dose to increase risk of death, as suggested by recent studies1. Although these comorbidities should be incorporated in survival analysis, such data is often missing from large datasets. We investigated whether calcifications, identified on the radiotherapy planning CT, can be used as a surrogate for general health. In particular, we investigated the interaction between calcifications, dose and survival. Methods Data from 846 unselected non-small cell lung cancer patients was used, all treated with 55 Gy in 20 fractions. Methodology was developed to automatically segment calcifications. The planning CT scan, acquired with a 3D protocol, and the lung and spinal cord delineations were processed using well-established image processing algorithms, e.g., convex hull, threshold, morphological operations, connected pixel analysis and flood filling to detect calcifications. A radiation oncologist manually identified calcifications in a small subset of 7 patients, and this group was used to estimate the success and error rate of the automatic segmentation. To reduce the impact of erroneous segmentations, an empirical threshold was used where patients showing more than 12 cm3 calcifications were excluded from analysis (n = 718 remaining). Finally, a multivariate Cox-regression analysis was performed for overall survival accounting for tumour size, total calcification volume and mean dose across all identified calcifications, including interaction between calcification volume and dose [1] . Results The success rate of the algorithm for identifying calcifications was 72.5%, its error rate was 23.2%. The Cox-regression analysis identified tumour size (continuous, p 0.0001) and the interaction of calcification volume and calcification dose (continuous, p = 0.023) as significant. The calcification volume (p = 0.172) or mean calcification dose alone (p = 0.535) were not found to be significant. Conclusions Multivariate analysis show a significant interaction between volume of the identified calcifications and radiotherapy dose predicting survival, which suggests that interaction rather than calcifications alone can be used as a factor in survival analysis. Further improvement and validation of our methodology is required, as well as linking our results with the established Agatston or Coronary Artery Calcium score. * EVO-FB share first authorship.

P176] Evaluating automatic contour propagation for estimating tumour shrinkage in small-cell lung cancer patients treated with chemo-radiotherapy

Purpose The clinical relevance of tumour shrinkage during thoracic radiotherapy remains unknown with studies showing conflicting results whether shrinkage predicts outcome. Previous work has relied on manual contouring on cone beam CT (CBCT). To enable large scale analysis, automated methods for contour propagation and quantification of tumour shrinkage are essential. Methods 20 small-cell lung cancer patients treated with concurrent chemo-radiotherapy were selected. All patients had a planning CT and, in total, 81 weekly CBCT scans were available. An experienced lung cancer radiation oncologist manually contoured the GTV on all scans. Additionally, the GTV contour of the planning CT was automatically propagated across the CBCTs, using ADMIRE (Research v1.13, Elekta AB, Stockholm, Sweden). To investigate the accuracy of the automated contour propagation, a comparison against the manual contours was performed using median distance to agreement (mDTA). To quantify tumour shrinkage, volumes were calculated for propagated and manual contours. Differences in volumes were assessed using a Bland–Altman plot. Finally, using the day one CBCT as reference volume, the relative percentage change was calculated at each time-point and fitted using linear regression. The slopes for manual versus propagated contours were compared. Results ADMIRE successfully propagated GTV contours all CBCTs. Concordance between the manual and propagated contours showed an unsigned mDTA of 3 mm across all CBCT. Results were worse for GTV contours overlapping with the mediastinum. The Bland–Altman analysis for absolute differences in volume at each time-point resulted a mean of −1.6 cm3. Relative differences provided a mean of −4.8%, with manual contours showing smaller volumes. Finally, manual contours estimated a consistently faster rate of tumour shrinkage (0.018 cm3/day) than automatically propagated contours (0.014 cm3/day). A linear fit between the manual and propagated gradients per patient found a slope of 0.68 (1 for perfect agreement). Conclusions The results show that automatic contour propagation performance is acceptable for use in wider research studies. There are differences between the manual and propagated contours, with the clinician estimated a larger rate of tumour shrinkage, which is suggestive of an observer bias. We will now apply this technique to investigate tumour shrinkage and outcome in the CONVERT trial patients.

P295] Reliability of planning dose distribution for assessment of dose distribution in rectum for prostate cancer

Purpose To assess the influence of changes of the rectum during radiotherapy on the dose distribution in rectum. Methods For 9 patients computerized 3D/IMRT CT plans were prepared. Before each session of radiotherapy the actual CT performed with the CT on rails was made. Due to technical problems on some days the CT was not made. In total, they were made 213 CTs. The lacking CTs were generated with the bootstrap method. The contour of rectum were was delineated according to RTOG recommendations. Total dose of 78 Gy in 2 Gy fraction dose was delivered. The actual CT images were transferred to Pinnacle treatment planning system. After fusion (on prostate) of actual and planning CT the actual dose distribution for each fraction was calculated. Next cumulative DVH and DVH for planning and the planning and actual no 1, 1-2, 1-3 CTs were calculated. For each set of data the NTCP was calculated. The model proposed by Defraene (Red, 2012) was used. The cumulative, the planning and actual no 1, 1-2, 1-3 and the planning DVHs were compared. Also we compared the planning volume and the cumulative rectum volume and the NTCPs. Results For some patients quite small changes of volume were observed, for some other quite large, e.g. the planning volume was 35.9 cm3, while cumulative one was 149.0 cm3. In 6 cases the cumulative NTCP was larger than the planning one. For two patients the differences were −0.11, 0.10 (NTCPcum-NTCPplan). For other patients the differences were smaller than ±0.05. Using the CTs made in the first few fractions enabled better estimation of the cumulative DVH but the improvements is not spectacular. Conclusions The changes of the rectum volume and position for some patients may influence quite large on DVH in rectum and as a result of this on NTCP. It is not clear for which patients such large differences may be expected.

Towards abdominal MRI-based treatment planning using population-based Hounsfield units for bulk density assignment

This study investigates the dosimetric impact of using population-based Hounsfield units (HUs) and ICRU-based HUs as a function of the number of tissue segments for bulk density assignment toward MRI-based treatment planning in the abdomen. To avoid potential geometric differences between CT and MR images, CT images rather than MR images were chosen to simulate an MRI-only planning scenario. A retrospective study was performed utilizing 18 patients that had previously undergone stereotactic body radiation therapy for liver or pancreas cancer. HU values in the CT datasets were collected for various tissue types, and compared with the HUs derived from ICRU report 46. Doses were recalculated using the fluence obtained from clinical plans and with (1) homogeneous assignment, (2) ICRU-based HU assignment and (3) population-based HU assignment using three, four, five, nine or ten tissue segments. Dose-volume metrics for targets and organs-at-risk for all scenarios were compared with those obtained using the clinical CT. For the planning target volume (PTV) D99.9%, the mean differences from clinical CT plans were −2.1% ± 3.9%, −0.6% ± 0.3% and −0.1% ± 0.3% for homogeneity, ICRU-HUs and population-HUs using ten tissue segments, respectively. The population-HU method resulted in better dosimetric accuracy compared to the ICRU-HU method (p-value < 0.05). The dosimetric accuracy of homogeneity plans was comparable to that of both ICRU-HU and population-HU plans when targets were far from the lungs but deteriorated when targets were close to the lungs. As the number of tissue segments decreased, the dosimetric accuracy for PTV D99.9% reduced for the population-HU method, from −0.1% for ten tissue segments to −0.4% for three tissue segments, while no such dependence was observed for the ICRU-HU method. Hence, to generate a clinically acceptable plan when using MRI to synthesize CT in the abdomen for treatment planning, it might be sufficient for electron density assignment with either the population-HU or ICRU-HU method to only use three tissue segments.

Comparison of computed tomography with magnetic resonance imaging for imaging-based clinical target volume contours in cervical cancer brachytherapy

PurposeTo compare CT- and MRI-based brachytherapy (BT) target volumes for patients with advanced cervical cancer so as to identify those who benefit most from MRI-based planning. We also studied how the natural mobility of the organ at risks (OARs) affects the given doses. Methods and MaterialsSubjects were 60 patients with International Federation of Gynecology and Obstetrics (FIGO) Stage IB–IVA cervical cancer. The CT high-risk clinical target volume (HR-CTV) was first delineated, then the MRI HR-CTV, with volume discrepancies calculated by subtraction. The DICE coefficient (DC) of similarity was calculated from a superimposition of the volumes. Maximum doses delivered to D2cc of OARs in CT and MRI plans were compared; the effect of time on the natural mobility was analyzed. ResultsThe mean HR-CTVs and the maximum doses given to OARs in CT- and MRI-based planes were similar. Multivariate analysis showed that deep infiltration affecting the uterine corpus and bowel loops adjacent to the cervix were the factors significantly impacting on the volume discrepancy between CT and MRI HR-CTV (p = 0.001, p = 0.045) and on the DC (p = 0.005, p = 0.028). Univariate analysis demonstrated that the FIGO stage had a significant impact on DC (p = 0.022). Patients with bowel loops adjacent to the cervix had lower body mass indices (p = 0.003). The median difference between the doses given in CT- and MRI-based plans, caused by mobility, were 0.5 Gy, 0.3 Gy, and 0.45 Gy per fraction for the rectum, bladder and sigmoid, respectively. No correlation of observed uncertainties and time between image acquisitions was detected. ConclusionsCT- or MRI-based scans at BT are adequate for OAR dose–volume histograms analysis. Cervical cancer patients with deep infiltration affecting the uterine corpus, a low body mass index with bowel loops adjacent to the cervix and an FIGO Stage III–IVA benefit most from MRI-based planning of BT.

Region-adaptive Deformable Registration of CT/MRI Pelvic Images via Learning-based Image Synthesis.

Registration of pelvic CT and MRI is highly desired as it can facilitate effective fusion of two modalities for prostate cancer radiation therapy, i.e., using CT for dose planning and MRI for accurate organ delineation. However, due to the large inter-modality appearance gaps and the high shape/appearance variations of pelvic organs, the pelvic CT/MRI registration is highly challenging. In this paper, we propose a region-adaptive deformable registration method for multi-modal pelvic image registration. Specifically, to handle the large appearance gaps, we first perform both CT-to-MRI and MRI-to-CT image synthesis by multi-target regression forest (MT-RF). Then, to use the complementary anatomical information in the two modalities for steering the registration, we select key points automatically from both modalities and use them together for guiding correspondence detection in the region-adaptive fashion. That is, we mainly use CT to establish correspondences for bone regions, and use MRI to establish correspondences for soft tissue regions. The number of key points is increased gradually during the registration, to hierarchically guide the symmetric estimation of the deformation fields. Experiments for both intra-subject and inter-subject deformable registration show improved performances compared with state-of-the-art multi-modal registration methods, which demonstrate the potentials of our method to be applied for the routine prostate cancer radiation therapy.