Evaluation of the effectiveness of novel single-intervention adaptive radiotherapy strategies based on daily dose accumulation.

Parotid gland (PG) shrinkage and neck volume reduction during radiotherapy of head and neck (H&N) cancer patients is a clinical issue that has prompted interest in adaptive radiotherapy (ART). This study focuses on the difference between planned dose and delivered dose and the possible effects of an efficient replanning strategy during the course of treatment. Six patients with H&N cancer treated by tomotherapy were retrospectively enrolled. Thirty daily dose distributions (DMVCT) were calculated on pretreatment megavoltage computed tomography (MVCT) scans. Deformable Image Registration which matched daily MVCT with treatment planning kilovoltage computed tomography was performed. Using the resulting deformation vector field, all daily DMVCT were deformed to the planning kilovoltage computed tomography and resulting doses were accumulated voxel per voxel. Cumulative DMVCT was compared to planned dose distribution performing γ-analysis (2 mm, 2% of 2.2 Gy). Two single-intervention ART strategies were executed on the 18th fraction whose previous data had suggested to be a suitable timepoint for a single replanning intervention: (1) replanning on the original target and deformed organ at risks (OARs) (a "safer" approach regarding tumor coverage) and (2) replanning on both deformed target and deformed OARs. DMVCT showed differences between planned and delivered doses (3D-γ 2mm/2%-passing rate = 85 ± 1%, p < 0.001). Voxel by voxel dose accumulation showed an increase in average dose of warped PG of 3.0 Gy ± 3.3 Gy. With ART the average dose of warped PG decreased by 3.2 Gy ± 1.7 Gy in comparison to delivered dose without replanning when both target and OARs were deformed. Average dose of warped PG decreased by 2.0 Gy ± 1.4 Gy when only OARs were deformed. Anatomical variations lead to increased doses to PGs. Efficient single-intervention ART-strategies with replanning on the 18th MVCT result a reduced PG dose. A strategy with deformation of both target and OAR resulted in the lowest PG dose, while formally maintaining PTV coverage. Deformation of only OAR nevertheless reduces PG dose and has less uncertainties regarding PTV coverage.

Expert system classifier for adaptive radiation therapy in prostate cancer.

A classifier-based expert system was developed to compare delivered and planned radiation therapy in prostate cancer patients. Its aim is to automatically identify patients that can benefit from an adaptive treatment strategy. The study predominantly addresses dosimetric uncertainties and critical issues caused by motion of hollow organs. 1200 MVCT images of 38 prostate adenocarcinoma cases were analyzed. An automatic daily re-contouring of structures (i.e. rectum, bladder and femoral heads), rigid/deformable registration and dose warping was carried out to simulate dose and volume variations during therapy. Support vector machine, K-means clustering algorithms and similarity index analysis were used to create an unsupervised predictive tool to detect incorrect setup and/or morphological changes as a consequence of inadequate patient preparation due to stochastic physiological changes, supporting clinical decision-making. After training on a dataset that was considered sufficiently dosimetrically stable, the system identified two equally sized macro clusters with distinctly different volumetric and dosimetric baseline properties and defined thresholds for these two clusters. Application to the test cohort resulted in 25% of the patients located outside the two macro clusters thresholds and which were therefore suspected to be dosimetrically unstable. In these patients, over the treatment course, mean volumetric changes of 30 and 40% for rectum and bladder were detected which possibly represents values justifying adjustment of patient preparation, frequent re-planning or a plan-of-the-day strategy. Based on our research, by combining daily IGRT images with rigid/deformable registration and dose warping, it is possible to apply a machine learning approach to the clinical setting obtaining useful information for a decision regarding an individualized adaptive strategy. Especially for treatments influenced by the movement of hollow organs, this could reduce inadequate treatments and possibly reduce toxicity, thereby increasing overall RT efficacy.

SIS epidemiological model for adaptive RT: Forecasting the parotid glands shrinkage during tomotherapy treatment.

PURPOSE: A susceptible-infected-susceptible (SIS) epidemic model was applied to radiation therapy (RT) treatments to predict morphological variations in head and neck (H&N) anatomy. METHODS: 360 daily MVCT images of 12 H&N patients treated by tomotherapy were analyzed in this retrospective study. Deformable image registration (DIR) algorithms, mesh grids, and structure recontouring, implemented in the RayStation treatment planning system (TPS), were applied to assess the daily organ warping. The parotid's warping was evaluated using the epidemiological approach considering each vertex as a single subject and its deformed vector field (DVF) as an infection. Dedicated IronPython scripts were developed to export daily coordinates and displacements of the region of interest (ROI) from the TPS. matlab tools were implemented to simulate the SIS modeling. Finally, the fully trained model was applied to a new patient. RESULTS: A QUASAR phantom was used to validate the model. The patients' validation was obtained setting 0.4 cm of vertex displacement as threshold and splitting susceptible (S) and infectious (I) cases. The correlation between the epidemiological model and the parotids' trend for further optimization of alpha and beta was carried out by Euclidean and dynamic time warping (DTW) distances. The best fit with experimental conditions across all patients (Euclidean distance of 4.09 ± 1.12 and DTW distance of 2.39 ± 0.66) was obtained setting the contact rate at 7.55 ± 0.69 and the recovery rate at 2.45 ± 0.26; birth rate was disregarded in this constant population. CONCLUSIONS: Combining an epidemiological model with adaptive RT (ART), the authors' novel approach could support image-guided radiation therapy (IGRT) to validate daily setup and to forecast anatomical variations. The SIS-ART model developed could support clinical decisions in order to optimize timing of replanning achieving personalized treatments.

Reducing absorbed dose to eye lenses in head CT examinations: the effect of bismuth shielding.

The eye lens is considered to be among the most radiosensitive human tissues. Brain CT scans may unnecessarily expose it to radiation even if the area of clinical interest is far from the eyes. The aim of this study is to implement a bismuth eye lens shielding system for Head-CT acquisitions in these cases. The study is focused on the assessment of the dosimetric characteristics of the shielding system as well as on its effect on image quality. The shielding system was tested in two set-ups which differ for distance ("contact" and "4 cm" Set up respectively). Scans were performed on a CTDI phantom and an anthropomorphic phantom. A reference set up without shielding system was acquired to establish a baseline. Image quality was assessed by signal (not HU converted), noise and contrast-to-noise ratio (CNR) evaluation. The overall dose reduction was evaluated by measuring the CTDIvol while the eye lens dose reduction was assessed by placing thermoluminescent dosimeters (TLDs) on an anthropomorphic phantom. The image quality analysis exhibits the presence of an artefact that mildly increases the CT number up to 3 cm below the shielding system. Below the artefact, the difference of the Signal and the CNR are negligible between the three different set-ups. Regarding the CTDI, the analysis demonstrates a decrease by almost 12 % (in the "contact" set-up) and 9 % (in the "4 cm" set-up). TLD measurements exhibit an eye lens dose reduction by 28.5 ± 5 and 21.1 ± 5 % respectively at the "contact" and the "4 cm" distance. No relevant artefact was found and image quality was not affected by the shielding system. Significant dose reductions were measured. These features make the shielding set-up useful for clinical implementation in both studied positions.

Quality assurance multicenter comparison of different MR scanners for quantitative diffusion-weighted imaging.

PURPOSE: To propose a magnetic resonance imaging (MRI) quality assurance procedure that can be used for multicenter comparison of different MR scanners for quantitative diffusion-weighted imaging (DWI). MATERIALS AND METHODS: Twenty-six centers (35 MR scanners with field strengths: 1T, 1.5T, and 3T) were enrolled in the study. Two different DWI acquisition series (b-value ranges 0-1000 and 0-3000 s/mm(2) , respectively) were performed for each MR scanner. All DWI acquisitions were performed by using a cylindrical doped water phantom. Mean apparent diffusion coefficient (ADC) values as well as ADC values along each of the three main orthogonal directions of the diffusion gradients (x, y, and z) were calculated. Short-term repeatability of ADC measurement was evaluated for 26 MR scanners. RESULTS: A good agreement was found between the nominal and measured mean ADC over all the centers. More than 80% of mean ADC measurements were within 5% from the nominal value, and the highest deviation and overall standard deviation were 9.3% and 3.5%, respectively. Short-term repeatability of ADC measurement was found <2.5% for all MR scanners. CONCLUSION: A specific and widely accepted protocol for quality controls in DWI is still lacking. The DWI quality assurance protocol proposed in this study can be applied in order to assess the reliability of DWI-derived indices before tackling single- as well as multicenter studies.

Radiosurgery Using Tomotherapy for Patients with Brain Oligo-metastasis: A Retrospective Analysis on Feasibility and Tolerance.

AIM: Aim of the study was to evaluate feasibility and toxicities of exclusive radiosurgery using tomotherapy in patients with brain oligo-metastases. PATIENTS AND METHODS: Between 2008 and 2013 68 patients underwent stereotactic radiosurgery (SRS). Mean patient age was 63 years. Brain was the only site involved in 32 patients, while 36 had extracranial disease. Pre-SRS MRI 56 patients had sovratentorial lesions, 10 subtentorial and 2 patients had both. Fifty-two patients had 1 brain lesion, 11 had 2, and 5 patients had three. All patients underwent SRS using Tomotherapy. The median delivered dose was 18 Gy. RESULTS: After a mean follow-up of 13 months, 14 patients were alive, while 54 patients had died. Two patients had complete response, 32 had partial response, 21 stable disease and 13 disease progression. Overall response rate was 80.9%. One- and two-year overall survival were 41,2% and 24,7%, while local control 61.5% and 37.7%. Toxicity was acceptable. CONCLUSION: SRS using tomotherapy has been proven feasible as non-invasive exclusive treatment for oligometastatic patients with good prognostic score.

A support vector machine tool for adaptive tomotherapy treatments: Prediction of head and neck patients criticalities.

PURPOSE: Adaptive radiation therapy (ART) is an advanced field of radiation oncology. Image-guided radiation therapy (IGRT) methods can support daily setup and assess anatomical variations during therapy, which could prevent incorrect dose distribution and unexpected toxicities. A re-planning to correct these anatomical variations should be done daily/weekly, but to be applicable to a large number of patients, still require time consumption and resources. Using unsupervised machine learning on retrospective data, we have developed a predictive network, to identify patients that would benefit of a re-planning. METHODS: 1200 MVCT of 40 head and neck (H&N) cases were re-contoured, automatically, using deformable hybrid registration and structures mapping. Deformable algorithm and MATLAB(®) homemade machine learning process, developed, allow prediction of criticalities for Tomotherapy treatments. RESULTS: Using retrospective analysis of H&N treatments, we have investigated and predicted tumor shrinkage and organ at risk (OAR) deformations. Support vector machine (SVM) and cluster analysis have identified cases or treatment sessions with potential criticalities, based on dose and volume discrepancies between fractions. During 1st weeks of treatment, 84% of patients shown an output comparable to average standard radiation treatment behavior. Starting from the 4th week, significant morpho-dosimetric changes affect 77% of patients, suggesting need for re-planning. The comparison of treatment delivered and ART simulation was carried out with receiver operating characteristic (ROC) curves, showing monotonous increase of ROC area. CONCLUSIONS: Warping methods, supported by daily image analysis and predictive tools, can improve personalization and monitoring of each treatment, thereby minimizing anatomic and dosimetric divergences from initial constraints.