Patrick V. Granton

Radiomics: Extracting more information from medical images using advanced feature analysis

Solid cancers are spatially and temporally heterogeneous. This limits the use of invasive biopsy based molecular assays but gives huge potential for medical imaging, which has the ability to capture intra-tumoural heterogeneity in a non-invasive way. During the past decades, medical imaging innovations with new hardware, new imaging agents and standardised protocols, allows the field to move towards quantitative imaging. Therefore, also the development of automated and reproducible analysis methodologies to extract more information from image-based features is a requirement. Radiomics--the high-throughput extraction of large amounts of image features from radiographic images--addresses this problem and is one of the approaches that hold great promises but need further validation in multi-centric settings and in the laboratory.

Small animal radiotherapy research platforms

Advances in conformal radiation therapy and advancements in pre-clinical radiotherapy research have recently stimulated the development of precise micro-irradiators for small animals such as mice and rats. These devices are often kilovolt x-ray radiation sources combined with high-resolution CT imaging equipment for image guidance, as the latter allows precise and accurate beam positioning. This is similar to modern human radiotherapy practice. These devices are considered a major step forward compared to the current standard of animal experimentation in cancer radiobiology research. The availability of this novel equipment enables a wide variety of pre-clinical experiments on the synergy of radiation with other therapies, complex radiation schemes, sub-target boost studies, hypofractionated radiotherapy, contrast-enhanced radiotherapy and studies of relative biological effectiveness, to name just a few examples. In this review we discuss the required irradiation and imaging capabilities of small animal radiation research platforms. We describe the need for improved small animal radiotherapy research and highlight pioneering efforts, some of which led recently to commercially available prototypes. From this, it will be clear that much further development is still needed, on both the irradiation side and imaging side. We discuss at length the need for improved treatment planning tools for small animal platforms, and the current lack of a standard therein. Finally, we mention some recent experimental work using the early animal radiation research platforms, and the potential they offer for advancing radiobiology research.

Extracting atomic numbers and electron densities from a dual source dual energy CT scanner: experiments and a simulation model.

BACKGROUND AND PURPOSE Dual energy CT (DECT) imaging can provide both the electron density ρ(e) and effective atomic number Z(eff), thus facilitating tissue type identification. This paper investigates the accuracy of a dual source DECT scanner by means of measurements and simulations. Previous simulation work suggested improved Monte Carlo dose calculation accuracy when compared to single energy CT for low energy photon brachytherapy, but lacked validation. As such, we aim to validate our DECT simulation model in this work. MATERIALS AND METHODS A cylindrical phantom containing tissue mimicking inserts was scanned with a second generation dual source scanner (SOMATOM Definition FLASH) to obtain Z(eff) and ρ(e). A model of the scanner was designed in ImaSim, a CT simulation program, and was used to simulate the experiment. RESULTS Accuracy of measured Z(eff) (labelled Z) was found to vary from -10% to 10% from low to high Z tissue substitutes while the accuracy on ρ(e) from DECT was about 2.5%. Our simulation reproduced the experiments within ±5% for both Z and ρ(e). CONCLUSIONS A clinical DECT scanner was able to extract Z and ρ(e) of tissue substitutes. Our simulation tool replicates the experiments within a reasonable accuracy.

Development and validation of a treatment planning system for small animal radiotherapy: SmART-Plan

BACKGROUND AND PURPOSE Image-guided equipment for precision irradiation of small animals for pre-clinical radiotherapy research became recently available. To enable downscaled radiotherapy studies that can be translated into human radiotherapy knowledge, a treatment planning system for pre-clinical studies is required. MATERIAL AND METHODS A dedicated treatment planning system (SmART-Plan) for small animal radiotherapy studies was developed. It is based on Monte Carlo simulation of particle transport in an animal. The voxel geometry is derived from the onboard cone beam CT imaging panel. SmART-Plan was validated using radiochromic film (RCF) dosimetry in various phantoms: uniform, multislab and a realistic plasticized mouse geometry. RESULTS Good agreement was obtained between SmART-Plan dose calculations and RCF dose measurements in all phantoms. For various delivered plans agreement was obtained within 10% for the majority of the targeted dose region, with larger differences between 10% and 20% near the penumbra regions and for the smallest 1mm collimator. Absolute depth and lateral dose distributions showed better agreement for 5 and 15-mm collimators than for a 1-mm collimator, indicating that accurate dose prediction for the smallest field sizes is difficult. CONCLUSION SmART-Plan offers a useful dose calculation tool for pre-clinical small animal irradiation studies.

Evaluation of a novel triple-channel radiochromic film analysis procedure using EBT2

A novel approach to read out radiochromic film was introduced recently by the manufacturer of GafChromic film. In this study, the performance of this triple-channel film dosimetry method was compared against the conventional single-red-channel film dosimetry procedure, with and without inclusion of a pre-irradiation (pre-IR) film scan, using EBT2 film and kilo- and megavoltage photon beams up to 10 Gy. When considering regions of interest averaged doses, the triple-channel method and both single-channel methods produced equivalent results. Absolute dose discrepancies between the triple-channel method, both single-channel methods and the treatment planning system calculated dose values, were no larger than 5 cGy for dose levels up to 2.2 Gy. Signal to noise in triple-channel dose images was found to be similar to signal to noise in single-channel dose images. The accuracy of resulting dose images from the triple- and single-channel methods with inclusion of pre-IR film scan was found to be similar. Results of a comparison of EBT2 data from a kilovoltage depth dose experiment to corresponding Monte Carlo depth dose data produced dose discrepancies of 9.5 ± 12 cGy and 7.6 ± 6 cGy for the single-channel method with inclusion of a pre-IR film scan and the triple-channel method, respectively. EBT2 showed to be energy sensitive at low kilovoltage energies with response differences of 11.9% and 15.6% in the red channel at 2 Gy between 50-225 kVp and 80-225 kVp photon spectra, respectively. We observed that the triple-channel method resulted in non-uniformity corrections of ±1% and consistency values of 0-3 cGy for the batches and dose levels studied. Results of this study indicate that the triple-channel radiochromic film read-out method performs at least as well as the single-channel method with inclusion of a pre-IR film scan, reduces film non-uniformity and saves time with elimination of a pre-IR film scan.

Simulation study on potential accuracy gains from dual energy CT tissue segmentation for low-energy brachytherapy Monte Carlo dose calculations

This work compares Monte Carlo (MC) dose calculations for (125)I and (103)Pd low-dose rate (LDR) brachytherapy sources performed in virtual phantoms containing a series of human soft tissues of interest for brachytherapy. The geometries are segmented (tissue type and density assignment) based on simulated single energy computed tomography (SECT) and dual energy (DECT) images, as well as the all-water TG-43 approach. Accuracy is evaluated by comparison to a reference MC dose calculation performed in the same phantoms, where each voxels material properties are assigned with exactly known values. The objective is to assess potential dose calculation accuracy gains from DECT. A CT imaging simulation package, ImaSim, is used to generate CT images of calibration and dose calculation phantoms at 80, 120, and 140 kVp. From the high and low energy images electron density ρ(e) and atomic number Z are obtained using a DECT algorithm. Following a correction derived from scans of the calibration phantom, accuracy on Z and ρ(e) of ±1% is obtained for all soft tissues with atomic number Z ∊ [6,8] except lung. GEANT4 MC dose calculations based on DECT segmentation agreed with the reference within ±4% for (103)Pd, the most sensitive source to tissue misassignments. SECT segmentation with three tissue bins as well as the TG-43 approach showed inferior accuracy with errors of up to 20%. Using seven tissue bins in our SECT segmentation brought errors within ±10% for (103)Pd. In general (125)I dose calculations showed higher accuracy than (103)Pd. Simulated image noise was found to decrease DECT accuracy by 3-4%. Our findings suggest that DECT-based segmentation yields improved accuracy when compared to SECT segmentation with seven tissue bins in LDR brachytherapy dose calculation for the specific case of our non-anthropomorphic phantom. The validity of our conclusions for clinical geometry as well as the importance of image noise in the tissue segmentation procedure deserves further experimental investigation.

A review of treatment planning for precision image-guided photon beam pre-clinical animal radiation studies

Recently, precision irradiators integrated with a high-resolution CT imaging device became available for pre-clinical studies. These research platforms offer significant advantages over older generations of animal irradiators in terms of precision and accuracy of image-guided radiation targeting. These platforms are expected to play a significant role in defining experiments that will allow translation of research findings to the human clinical setting. In the field of radiotherapy, but also others such as neurology, the platforms create unique opportunities to explore e.g. the synergy between radiation and drugs or other agents. To fully exploit the advantages of this new technology, accurate methods are needed to plan the irradiation and to calculate the three-dimensional radiation dose distribution in the specimen. To this end, dedicated treatment planning systems are needed. In this review we will discuss specific issues for precision irradiation of small animals, we will describe the workflow of animal treatment planning, and we will examine several dose calculation algorithms (factorization, superposition-convolution, Monte Carlo simulation) used for animal irradiation with kilovolt photon beams. Issues such as dose reporting methods, photon scatter, tissue segmentation and motion will also be discussed briefly.

A combined dose calculation and verification method for a small animal precision irradiator based on onboard imaging

PURPOSE Novel small animal precision microirradiators (micro-IR) are becoming available for preclinical use and are often equipped with onboard imaging (OBI) devices. We investigated the use of OBI as a means to infer the accuracy of the delivered treatment plan. METHODS Monte Carlo modeling of the micro-IR including an elliptical Gaussian electron beam incident on the x-ray tube was used to score dose and to continue photon transport to the plane of the OBI device. A model of the OBI detector response was used to generate simulated onboard images. Experimental OBI was performed at 225 kVp, gain∕offset and scatter-glare were corrected. Simulated and experimentally obtained onboard images of phantoms and a mouse specimen were compared for a range of photon beam sizes from 2.5 cm down to 0.1 cm. RESULTS Simulated OBI can be used in small animal radiotherapy to determine if a treatment plan was delivered according to the prescription within an uncertainty of 5% for beams as small as 4 mm in diameter. For collimated beams smaller than 4 mm, beam profile differences remain primarily in the penumbra region of the smallest beams, which may be tolerable for specific preclinical micro-IR investigations. CONCLUSIONS Comparing simulated to acquired OBI during small animal treatment radiotherapy represents a useful treatment delivery tool.PURPOSE Novel small animal precision microirradiators (micro-IR) are becoming available for preclinical use and are often equipped with onboard imaging (OBI) devices. We investigated the use of OBI as a means to infer the accuracy of the delivered treatment plan. METHODS Monte Carlo modeling of the micro-IR including an elliptical Gaussian electron beam incident on the x-ray tube was used to score dose and to continue photon transport to the plane of the OBI device. A model of the OBI detector response was used to generate simulated onboard images. Experimental OBI was performed at 225 kVp, gain/offset and scatter-glare were corrected. Simulated and experimentally obtained onboard images of phantoms and a mouse specimen were compared for a range of photon beam sizes from 2.5 cm down to 0.1 cm. RESULTS Simulated OBI can be used in small animal radiotherapy to determine if a treatment plan was delivered according to the prescription within an uncertainty of 5% for beams as small as 4 mm in diameter. For collimated beams smaller than 4 mm, beam profile differences remain primarily in the penumbra region of the smallest beams, which may be tolerable for specific preclinical micro-IR investigations. CONCLUSIONS Comparing simulated to acquired OBI during small animal treatment radiotherapy represents a useful treatment delivery tool.

A Longitudinal Evaluation of Partial Lung Irradiation in Mice by Using a Dedicated Image-Guided Small Animal Irradiator

PURPOSE In lung cancer radiation therapy, the dose constraints are determined mostly by healthy lung toxicity. Preclinical microirradiators are a new tool to evaluate treatment strategies closer to clinical irradiation devices. In this study, we quantified local changes in lung density symptomatic of radiation-induced lung fibrosis (RILF) after partial lung irradiation in mice by using a precision image-guided small animal irradiator integrated with micro-computed tomography (CT) imaging. METHODS AND MATERIALS C57BL/6 adult male mice (n=76) were divided into 6 groups: a control group (0 Gy) and groups irradiated with a single fraction of 4, 8, 12, 16, or 20 Gy using 5-mm circular parallel-opposed fields targeting the upper right lung. A Monte Carlo model of the small animal irradiator was used for dose calculations. Following irradiation, all mice were imaged at regular intervals over 39 weeks (10 time points total). Nonrigid deformation was used to register the initial micro-CT scan to all subsequent scans. RESULTS Significant differences could be observed between the 3 highest (>10 Gy) and 3 lowest irradiation (<10 Gy) dose levels. A mean difference of 120 ± 10 HU between the 0- and 20-Gy groups was observed at week 39. RILF was found to be spatially limited to the irradiated portion of the lung. CONCLUSIONS The data suggest that the severity of RILF in partial lung irradiation compared to large field irradiation in mice for the same dose is reduced, and therefore higher doses can be tolerated.

NOTCH blockade combined with radiation therapy and temozolomide prolongs survival of orthotopic glioblastoma

Glioblastoma multiforme (GBM) is the most common malignant brain tumor in adults. The current standard of care includes surgery followed by radiotherapy (RT) and chemotherapy with temozolomide (TMZ). Treatment often fails due to the radiation resistance and intrinsic or acquired TMZ resistance of a small percentage of cells with stem cell-like behavior (CSC). The NOTCH signaling pathway is expressed and active in human glioblastoma and NOTCH inhibitors attenuate tumor growth in vivo in xenograft models. Here we show using an image guided micro-CT and precision radiotherapy platform that a combination of the clinically approved NOTCH/γ-secretase inhibitor (GSI) RO4929097 with standard of care (TMZ + RT) reduces tumor growth and prolongs survival compared to dual combinations. We show that GSI in combination with RT and TMZ attenuates proliferation, decreases 3D spheroid growth and results into a marked reduction in clonogenic survival in primary and established glioma cell lines. We found that the glioma stem cell marker CD133, SOX2 and Nestin were reduced following combination treatments and NOTCH inhibitors albeit in a different manner. These findings indicate that NOTCH inhibition combined with standard of care treatment has an anti-glioma stem cell effect which provides an improved survival benefit for GBM and encourages further translational and clinical studies.