X-ray fluorescence computed tomography (XFCT) is a technique that can identify, quantify, and locate elements within objects by detecting x-ray fluorescence (characteristic x-rays) stimulated by an excitation source, typically derived from a synchrotron. However, the use of a synchrotron limits practicality and accessibility of XFCT for routine biomedical imaging applications. Therefore, we have developed the ability to perform XFCT on a benchtop setting with ordinary polychromatic x-ray sources. Here, we report our postmortem study that demonstrates the use of benchtop XFCT to accurately image the distribution of gold nanoparticles (GNPs) injected into a tumor-bearing mouse. The distribution of GNPs as determined by benchtop XFCT was validated using inductively coupled plasma mass spectrometry. This investigation shows drastically enhanced sensitivity and specificity of GNP detection and quantification with benchtop XFCT, up to two orders of magnitude better than conventional x-ray CT. The results also reaffirm the unique capabilities of benchtop XFCT for simultaneous determination of the spatial distribution and concentration of nonradioactive metallic probes, such as GNPs, within the context of small animal imaging. Overall, this investigation identifies a clear path toward in vivo molecular imaging using benchtop XFCT techniques in conjunction with GNPs and other metallic probes.

A conventional x-ray fluorescence computed tomography (XFCT) technique requires monochromatic synchrotron x-rays to simultaneously determine the spatial distribution and concentration of various elements such as metals in a sample. However, the synchrotron-based XFCT technique appears to be unsuitable for in vivo imaging under a typical laboratory setting. In this study we demonstrated, for the first time to our knowledge, the possibility of performing XFCT imaging of a small animal-sized object containing gold nanoparticles (GNPs) at relatively low concentrations using polychromatic diagnostic energy range x-rays. Specifically, we created a phantom made of polymethyl methacrylate plastic containing two cylindrical columns filled with saline solution at 1 and 2 wt% GNPs, respectively, mimicking tumors/organs within a small animal. XFCT scanning of the phantom was then performed using microfocus 110 kVp x-ray beam and cadmium telluride (CdTe) x-ray detector under a pencil beam geometry after proper filtering of the x-ray beam and collimation of the detector. The reconstructed images clearly identified the locations of the two GNP-filled columns with different contrast levels directly proportional to gold concentration levels. On the other hand, the current pencil-beam implementation of XFCT is not yet practical for routine in vivo imaging tasks with GNPs, especially in terms of scanning time. Nevertheless, with the use of multiple detectors and a limited number of projections, it may still be used to image some objects smaller than the current phantom size. The current investigation suggests several modification strategies of the current XFCT setup, such as the adoption of the quasi-monochromatic cone/fan x-ray beam and XFCT-specific spatial filters or pinhole detector collimators, in order to establish the ultimate feasibility of a bench-top XFCT system for GNP-based preclinical molecular imaging applications.

This report presents the first experimental demonstration, to our knowledge, of benchtop polychromatic cone-beam x-ray fluorescence computed tomography (XFCT) for a simultaneous determination of the spatial distribution and amount of gold nanoparticles (GNPs) within small-animal-sized objects. The current benchtop experimental setup successfully produced XFCT images accurately showing the regions containing small amount of GNPs (on the order of 0.1 mg) within a 3-cm diameter plastic phantom. In particular, the performance of the current XFCT setup was improved remarkably (e.g., at least a factor of 3 reduction in XFCT scan time) using a tin-filtered polychromatic beam in comparison with a lead-filtered beam. The results of this study strongly suggest the current benchtop XFCT configuration can be made practical with a few modifications such as the deployment of array detectors, while meeting realistic constraints on x-ray dose, scan time, and image resolution for routine pre-clinical in-vivo imaging with GNPs.

Magnetic resonance imaging (MRI)-only radiotherapy treatment planning is attractive since MRI provides superior soft tissue contrast without ionizing radiation compared with computed tomography (CT). However, it requires the generation of pseudo CT from MRI images for patient setup and dose calculation. Our machine-learning-based method to generate pseudo CT images has been shown to provide pseudo CT images with excellent image quality, while its dose calculation accuracy remains an open question. In this study, we aim to investigate the accuracy of dose calculation in brain frameless stereotactic radiosurgery (SRS) using pseudo CT images which are generated from MRI images using the machine learning-based method developed by our group. We retrospectively investigated a total of 19 treatment plans from 14 patients, each of whom has CT simulation and MRI images acquired during pretreatment. The dose distributions of the same treatment plans were calculated on original CT simulation images as ground truth, as well as on pseudo CT images generated from MRI images. Clinically-relevant DVH metrics and gamma analysis were extracted from both ground truth and pseudo CT results for comparison and evaluation. The side-by-side comparisons on image quality and dose distributions demonstrated very good agreement of image contrast and calculated dose between pseudo CT and original CT. The average differences in Dose-volume histogram (DVH) metrics for Planning target volume (PTVs) were less than 0.6%, and no differences in those for organs at risk at a significance level of 0.05. The average pass rate of gamma analysis was 99%. These quantitative results strongly indicate that the pseudo CT images created from MRI images using our proposed machine learning method are accurate enough to replace current CT simulation images for dose calculation in brain SRS treatment. This study also demonstrates the great potential for MRI to completely replace CT scans in the process of simulation and treatment planning.

Some investigators have shown tumor cell killing enhancement in vitro and tumor regression in mice associated with the loading of gold nanoparticles (GNPs) before proton treatments. Several Monte Carlo (MC) investigations have also demonstrated GNP-mediated proton dose enhancement. However, further studies need to be done to quantify the individual physical factors that contribute to the dose enhancement or cell-kill enhancement (or radiosensitization). Thus, the current study investigated the contributions of particle-induced x-ray emission (PIXE), particle-induced gamma-ray emission (PIGE), Auger and secondary electrons, and activation products towards the total dose enhancement. Specifically, GNP-mediated dose enhancement was measured using strips of radiochromic film that were inserted into vials of cylindrical GNPs, i.e. gold nanorods (GNRs), dispersed in a saline solution (0.3 mg of GNRs/g or 0.03% of GNRs by weight), as well as vials containing water only, before proton irradiation. MC simulations were also performed with the tool for particle simulation code using the film measurement setup. Additionally, a high-purity germanium detector system was used to measure the photon spectrum originating from activation products created from the interaction of protons and spherical GNPs present in a saline solution (20 mg of GNPs/g or 2% of GNPs by weight). The dose enhancement due to PIXE/PIGE recorded on the films in the GNR-loaded saline solution was less than the experimental uncertainty of the film dosimetry (<2%). MC simulations showed highly localized dose enhancement (up to a factor 17) in the immediate vicinity (<100 nm) of GNRs, compared with hypothetical water nanorods (WNRs), mostly due to GNR-originated Auger/secondary electrons; however, the average dose enhancement over the entire GNR-loaded vial was found to be minimal (0.1%). The dose enhancement due to the activation products from GNPs was minimal (<0.1%) as well. In conclusion, under the currently investigated conditions that are considered clinically relevant, PIXE, PIGE, and activation products contribute minimally to GNP/GNR-mediated proton dose enhancement, whereas Auger/secondary electrons contribute significantly but only at short distances (<100 nm) from GNPs/GNRs.