Multi-Institutional Stereotactic Body Radiation Therapy Incident Learning: Evaluation of Safety Barriers Using a Human Factors Analysis and Classification System.

OBJECTIVES: Stereotactic body radiation therapy (SBRT) can improve therapeutic ratios and patient convenience, but delivering higher doses per fraction increases the potential for patient harm. Incident learning systems (ILSs) are being increasingly adopted in radiation oncology to analyze reported events. This study used an ILS coupled with a Human Factor Analysis and Classification System (HFACS) and barriers management to investigate the origin and detection of SBRT events and to elucidate how safeguards can fail allowing errors to propagate through the treatment process. METHODS: Reported SBRT events were reviewed using an in-house ILS at 4 institutions over 2014-2019. Each institution used a customized care path describing their SBRT processes, including designated safeguards to prevent error propagation. Incidents were assigned a severity score based on the American Association of Physicists in Medicine Task Group Report 275. An HFACS system analyzed failing safeguards. RESULTS: One hundred sixty events were analyzed with 106 near misses (66.2%) and 54 incidents (33.8%). Fifty incidents were designated as low severity, with 4 considered medium severity. Incidents most often originated in the treatment planning stage (38.1%) and were caught during the pretreatment review and verification stage (37.5%) and treatment delivery stage (31.2%). An HFACS revealed that safeguard failures were attributed to human error (95.2%), routine violation (4.2%), and exceptional violation (0.5%) and driven by personnel factors 32.1% of the time, and operator condition also 32.1% of the time. CONCLUSIONS: Improving communication and documentation, reducing time pressures, distractions, and high workload should guide proposed improvements to safeguards in radiation oncology.

Irradiation of the Normal Murine Tongue Causes Upregulation and Activation of Transient Receptor Potential (TRP) Ion Channels.

Signal transduction at sensory neurons occurs via transmembrane flux of cations, which is largely governed by the transient receptor potential (TRP) family of ion channels. It is unknown whether TRP channel activation contributes to the pain that accompanies radiation-induced oral mucositis. This study sought to characterize changes in TRP channel expression and function that occur in the locally irradiated tissues and afferent neurons of mice. Female CD-1 mice received single high-dose (27 Gy) tongue irradiation, or sham irradiation. Animals were euthanized either before overt glossitis developed (days 1 and 5 postirradiation), when glossitis was severe (day 11), or after mice had recovered (days 21 and 45). Tongue irradiation caused upregulation of the Trpv1 gene in trigeminal ganglia (TG) neurons. Other TRP genes (Trpv2, Trpv4, Trpa1, Trpm8) and Gfrα3 (which acts upstream of several TRP channels) were also upregulated in TGs and/or tongue tissue, in response to radiation. Ex vivo calcium imaging experiments demonstrated that the proportions of TG neurons responding to histamine (an activator of TRPV1, TRPV4 and TRPA1), TNF-α (an activator of TRPV1, TRPV2 and TRPV4), and capsaicin (a TRPV1 agonist), were increased as early as one day after tongue irradiation; these changes persisted for at least 21 days. In a subsequent experiment, we found that genetic deletion of TRPV1 mitigated weight loss (a surrogate marker of pain severity) in mice with severe glossitis. The results intimate that various TRP channels, and TRPV1 in particular, should be explored as analgesic targets for patients experiencing pain after oral irradiation.

Implementation of total body photon irradiation as part of an institutional bone marrow transplant program for the treatment of canine lymphoma and leukemias.

A total body irradiation (TBI) protocol was developed to support a bone marrow transplant (BMT) program for the treatment of canine hematologic malignancies. The purpose of this prospective study is to describe implementation of the protocol and resultant dosimetry. Nongraphic manual treatment planning using 6 MV photons, isocentric delivery, 40 × 40 cm field size, wall-mounted lasers to verify positioning, a lucite beam spoiler (without use of bolus material), a dose rate of 8.75 cGy/min at patient isocenter, and a source-to-axis distance of 338 cm were used for TBI. A monitor unit calculation formula was derived using ion chamber measurements and a solid water phantom. Five thermoluminescent dosimeters (TLDs) were used at various anatomic locations in each of four cadaver dogs, to verify fidelity of the monitor unit formula prior to clinical implementation. In vivo dosimetric data were then collected with five TLDs at various anatomic locations in six patients treated with TBI. A total dose of 10 Gy divided into two 5 Gy fractions was delivered approximately 16 h apart, immediately followed by autologous stem cell transplant. The mean difference between prescribed and delivered doses ranged from 99% to 109% for various sites in cadavers, and from 83% to 121% in clinical patients. The mean total body dose in cadavers and clinical patients when whole body dose was estimated by averaging doses measured by variably placed TLDs ranged from 98% to 108% and 93% to 102% of the prescribed dose, respectively, which was considered acceptable. This protocol could be used for institutional implementation of TBI.

Nocifensive Behaviors in Mice with Radiation-Induced Oral Mucositis.

Oral mucositis can result in significant dysphagia, and is the most common dose-limiting acute toxicity in head and neck cancer patients receiving chemoradiotherapy. There is a critical need to determine the cellular and molecular mechanisms that underlie radiotherapy-associated discomfort in patients with mucositis. The objective was to induce oral mucositis in mice, using a clinical linear accelerator, and to quantify resultant discomfort, and characterize peripheral sensitization. A clinical linear accelerator was used to deliver ionizing radiation to the oral cavity of mice. Mucositis severity scoring, and various behavioral assays were performed to quantify bouts of orofacial wiping and scratching, bite force, gnawing behavior and burrowing activity. Calcium imaging was performed on neurons of the trigeminal ganglia. Glossitis was induced with a single fraction of at least 27 Gy. Body weight decreased and subsequently returned to baseline, in concert with development and resolution of mucositis, which was worst at day 10 and 11 postirradiation, however was resolved within another 10 days. Neither bite force, nor gnawing behavior were measurably affected. However, burrowing activity was decreased, and both facial wiping and scratching were increased while mice had visible mucositis lesions. Sensory nerves of irradiated mice were more responsive to histamine, tumor necrosis factor alpha and capsaicin. Radiation-induced glossitis is associated with hyper-reactivity of sensory neurons in the trigeminal ganglia of mice, and is accompanied by several behaviors indicative of both itch and pain. These data validate an appropriate model for cancer treatment related discomfort in humans.

Outcomes of Spatially Fractionated Radiotherapy (GRID) for Bulky Soft Tissue Sarcomas in a Large Animal Model.

GRID directs alternating regions of high- and low-dose radiation at tumors. A large animal model mimicking the geometries of human treatments is needed to complement existing rodent systems (eg, microbeam) and clarify the physical and biological attributes of GRID. A pilot study was undertaken in pet dogs with spontaneous soft tissue sarcomas to characterize responses to GRID. Subjects were treated with either 20 Gy (3 dogs) or 25 Gy (3 dogs), delivered using 6 MV X-rays and a commercial GRID collimator. Acute toxicity and tumor responses were assessed 2, 4, and 6 weeks later. Acute Radiation Therapy Oncology Group grade I skin toxicity was observed in 3 of the 6 dogs; none experienced a measurable response, per Response Evaluation Criteria in Solid Tumors. Serum vascular endothelial growth factor, tumor necrosis factor α, and secretory sphingomyelinase were assayed at baseline, 1, 4, 24, and 48 hours after treatment. There was a trend toward platelet-corrected serum vascular endothelial growth factor concentration being lower 1 and 48 hours after GRID than at baseline. There was a significant decrease in secretory sphingomyelinase activity 48 hours after 25 Gy GRID ( P = .03). Serum tumor necrosis factor α was quantified measurable at baseline in 4 of the 6 dogs and decreased in each of those subjects at all post-GRID time points. The new information generated by this study includes the observation that high-dose, single fraction application of GRID does not induce measurable reduction in volume of canine soft tissue sarcomas. In contrast to previously published data, these data suggest that GRID may be associated with at least short-term reduction in serum concentration of vascular endothelial growth factor and serum activity of secretory sphingomyelinase. Because GRID can be applied safely, and these tumors can be subsequently surgically resected as part of routine veterinary care, pet dogs with sarcomas are an appealing model for studying the radiobiologic responses to spatially fractionated radiotherapy.

DOSIMETRIC CONSEQUENCES OF USING CONTRAST-ENHANCED COMPUTED TOMOGRAPHIC IMAGES FOR INTENSITY-MODULATED STEREOTACTIC BODY RADIOTHERAPY PLANNING.

Potential benefits of planning radiation therapy on a contrast-enhanced computed tomography scan (ceCT) should be weighed against the possibility that this practice may be associated with an inadvertent risk of overdosing nearby normal tissues. This study investigated the influence of ceCT on intensity-modulated stereotactic body radiotherapy (IM-SBRT) planning. Dogs with head and neck, pelvic, or appendicular tumors were included in this retrospective cross-sectional study. All IM-SBRT plans were constructed on a pre- or ceCT. Contours for tumor and organs at risk (OAR) were manually constructed and copied onto both CT's; IM-SBRT plans were calculated on each CT in a manner that resulted in equal radiation fluence. The maximum and mean doses for OAR, and minimum, maximum, and mean doses for targets were compared. Data were collected from 40 dogs per anatomic site (head and neck, pelvis, and limbs). The average dose difference between minimum, maximum, and mean doses as calculated on pre- and ceCT plans for the gross tumor volume was less than 1% for all anatomic sites. Similarly, the differences between mean and maximum doses for OAR were less than 1%. The difference in dose distribution between plans made on CTs with and without contrast enhancement was tolerable at all treatment sites. Therefore, although caution would be recommended when planning IM-SBRT for tumors near "reservoirs" for contrast media (such as the heart and urinary bladder), findings supported the use of ceCT with this dose calculation algorithm for both target delineation and IM-SBRT treatment planning.

Monitor unit calculations for external photon and electron beams: Report of the AAPM Therapy Physics Committee Task Group No. 71.

A protocol is presented for the calculation of monitor units (MU) for photon and electron beams, delivered with and without beam modifiers, for constant source-surface distance (SSD) and source-axis distance (SAD) setups. This protocol was written by Task Group 71 of the Therapy Physics Committee of the American Association of Physicists in Medicine (AAPM) and has been formally approved by the AAPM for clinical use. The protocol defines the nomenclature for the dosimetric quantities used in these calculations, along with instructions for their determination and measurement. Calculations are made using the dose per MU under normalization conditions, D'0, that is determined for each user's photon and electron beams. For electron beams, the depth of normalization is taken to be the depth of maximum dose along the central axis for the same field incident on a water phantom at the same SSD, where D'0 = 1 cGy/MU. For photon beams, this task group recommends that a normalization depth of 10 cm be selected, where an energy-dependent D'0 ≤ 1 cGy/MU is required. This recommendation differs from the more common approach of a normalization depth of dm, with D'0 = 1 cGy/MU, although both systems are acceptable within the current protocol. For photon beams, the formalism includes the use of blocked fields, physical or dynamic wedges, and (static) multileaf collimation. No formalism is provided for intensity modulated radiation therapy calculations, although some general considerations and a review of current calculation techniques are included. For electron beams, the formalism provides for calculations at the standard and extended SSDs using either an effective SSD or an air-gap correction factor. Example tables and problems are included to illustrate the basic concepts within the presented formalism.

Automated classification of lung bronchovascular anatomy in CT using AdaBoost.

Lung CAD systems require the ability to classify a variety of pulmonary structures as part of the diagnostic process. The purpose of this work was to develop a methodology for fully automated voxel-by-voxel classification of airways, fissures, nodules, and vessels from chest CT images using a single feature set and classification method. Twenty-nine thin section CT scans were obtained from the Lung Image Database Consortium (LIDC). Multiple radiologists labeled voxels corresponding to the following structures: airways (trachea to 6th generation), major and minor lobar fissures, nodules, and vessels (hilum to peripheral), and normal lung parenchyma. The labeled data was used in conjunction with a supervised machine learning approach (AdaBoost) to train a set of ensemble classifiers. Each ensemble classifier was trained to detect voxels part of a specific structure (either airway, fissure, nodule, vessel, or parenchyma). The feature set consisted of voxel attenuation and a small number of features based on the eigenvalues of the Hessian matrix (used to differentiate structures by shape). When each ensemble classifier was composed of 20 weak classifiers, the AUC values for the airway, fissure, nodule, vessel, and parenchyma classifiers were 0.984+/-0.011, 0.949+/-0.009, 0.945+/-0.018, 0.953+/-0.016, and 0.931+/-0.015, respectively. The strong results suggest that this could be an effective input to higher-level anatomical based segmentation models with the potential to improve CAD performance.

Solitary pulmonary nodule diagnosis on CT: results of an observer study.

RATIONALE AND OBJECTIVES: To investigate the performance of observers with different levels of experience in distinguishing between benign and malignant solitary pulmonary nodules (SPN) on CT, and to determine the effects on interpretation of three different conditions: image data alone, the addition of clinical data, and the addition of output from a computer-aided diagnosis (CAD) system. MATERIALS AND METHODS: 28 thin-section CT datasets of SPNs with proven diagnoses (15 malignant and 13 benign) were used to measure observer performance. Readers were categorized according to their experience and read the cases in random order. For each case readers were asked to assign a level of confidence on a scale from 0.0-1.0 (0.0 benign, 1.0 malignant) for the diagnosis of the nodule. Each reader scored the cases based on review of image data alone (phase 1), then with limited clinical data (phase 2), and finally with CAD output (phase 3). To assess performance, multiple reader multiple case (MRMC) receiver operating characteristic (ROC) analysis was used. RESULTS: 2 thoracic radiologists, 1 thoracic radiology fellow, 2 nonthoracic radiologists, and 3 radiology residents read the cases. The average area under the ROC curve for all readers (A(z)) at each stage was 0.68, 0.75, and 0.81, for image data alone, with clinical data, and with CAD output respectively. The difference in performance between phases (2 and 3) and (1 and 3) was significantly different (P = 0.018 and P = 0.020). However, the difference between phases (1 and 2) was not significantly different (P = 0.155). CONCLUSION: Diagnostic performance increased significantly with the addition of CAD output. With further validation CAD output may play a significant role in SPN management.