Technical note: Minimizing CIED artifacts on a 0.35 T MRI-Linac using deep learning.

BACKGROUND: Artifacts from implantable cardioverter defibrillators (ICDs) are a challenge to magnetic resonance imaging (MRI)-guided radiotherapy (MRgRT). PURPOSE: This study tested an unsupervised generative adversarial network to mitigate ICD artifacts in balanced steady-state free precession (bSSFP) cine MRIs and improve image quality and tracking performance for MRgRT. METHODS: Fourteen healthy volunteers (Group A) were scanned on a 0.35 T MRI-Linac with and without an MR conditional ICD taped to their left pectoral to simulate an implanted ICD. bSSFP MRI data from 12 of the volunteers were used to train a CycleGAN model to reduce ICD artifacts. The data from the remaining two volunteers were used for testing. In addition, the dataset was reorganized three times using a Leave-One-Out scheme. Tracking metrics [Dice similarity coefficient (DSC), target registration error (TRE), and 95 percentile Hausdorff distance (95% HD)] were evaluated for whole-heart contours. Image quality metrics [normalized root mean square error (nRMSE), peak signal-to-noise ratio (PSNR), and multiscale structural similarity (MS-SSIM) scores] were evaluated. The technique was also tested qualitatively on three additional ICD datasets (Group B) including a patient with an implanted ICD. RESULTS: For the whole-heart contour with CycleGAN reconstruction: 1) Mean DSC rose from 0.910 to 0.935; 2) Mean TRE dropped from 4.488 to 2.877 mm; and 3) Mean 95% HD dropped from 10.236 to 7.700 mm. For the whole-body slice with CycleGAN reconstruction: 1) Mean nRMSE dropped from 0.644 to 0.420; 2) Mean MS-SSIM rose from 0.779 to 0.819; and 3) Mean PSNR rose from 18.744 to 22.368. The three Group B datasets evaluated qualitatively displayed a reduction in ICD artifacts in the heart. CONCLUSION: CycleGAN-generated reconstructions significantly improved both tracking and image quality metrics when used to mitigate artifacts from ICDs.

Feasibility of surface-guidance combined with CBCT for intra-fractional breath-hold motion management during Ethos RT.

PURPOSE: High-quality CBCT and AI-enhanced adaptive planning techniques allow CBCT-guided stereotactic adaptive radiotherapy (CT-STAR) to account for inter-fractional anatomic changes. Studies of intra-fractional respiratory motion management with a surface imaging solution for CT-STAR have not been fully conducted. We investigated intra-fractional motion management in breath-hold Ethos-based CT-STAR and CT-SBRT (stereotactic body non-adaptive radiotherapy) using optical surface imaging combined with onboard CBCTs. METHODS: Ten cancer patients with mobile lower lung or upper abdominal malignancies participated in an IRB-approved clinical trial (Phase I) of optical surface image-guided Ethos CT-STAR/SBRT. In the clinical trial, a pre-configured gating window (± 2 mm in AP direction) on optical surface imaging was used for manually triggering intra-fractional CBCT acquisition and treatment beam irradiation during breath-hold (seven patients for the end of exhalation and three patients for the end of inhalation). Two inter-fractional CBCTs at the ends of exhalation and inhalation in each fraction were acquired to verify the primary direction and range of the tumor/imaging-surrogate (donut-shaped fiducial) motion. Intra-fractional CBCTs were used to quantify the residual motion of the tumor/imaging-surrogate within the pre-configured breath-hold window in the AP direction. Fifty fractions of Ethos RT were delivered under surface image-guidance: Thirty-two fractions with CT-STAR (adaptive RT) and 18 fractions with CT-SBRT (non-adaptive RT). The residual motion of the tumor was quantified by determining variations in the tumor centroid position. The dosimetric impact on target coverage was calculated based on the residual motion. RESULTS: We used 46 fractions for the analysis of intra-fractional residual motion and 43 fractions for the inter-fractional motion analysis due to study constraints. Using the image registration method, 43 pairs of inter-fractional CBCTs and 100 intra-fractional CBCTs attached to dose maps were analyzed. In the motion range study (image registration) from the inter-fractional CBCTs, the primary motion (mean ± std) was 16.6 ± 9.2 mm in the SI direction (magnitude: 26.4 ± 11.3 mm) for the tumors and 15.5 ± 7.3 mm in the AP direction (magnitude: 20.4 ± 7.0 mm) for the imaging-surrogate, respectively. The residual motion of the tumor (image registration) from intra-fractional breath-hold CBCTs was 2.2 ± 2.0 mm for SI, 1.4 ± 1.4 mm for RL, and 1.3 ± 1.3 mm for AP directions (magnitude: 3.5 ± 2.1 mm). The ratio of the actual dose coverage to 99%, 90%, and 50% of the target volume decreased by 0.95 ± 0.11, 0.96 ± 0.10, 0.99 ± 0.05, respectively. The mean percentage of the target volume covered by the prescribed dose decreased by 2.8 ± 4.4%. CONCLUSION: We demonstrated the intra-fractional motion-managed treatment strategy in breath-hold Ethos CT-STAR/SBRT using optical surface imaging and CBCT. While the controlled residual tumor motion measured at 3.5 mm exceeded the predetermined setup value of 2 mm, it is important to note that this motion still fell within the clinically acceptable range defined by the PTV margin of 5 mm. Nonetheless, additional caution is needed with intra-fractional motion management in breath-hold Ethos CT-STAR/SBRT using optical surface imaging and CBCT.

Clinical application of a template-guided automated planning routine.

PURPOSE: Determine the dosimetric quality and the planning time reduction when utilizing a template-based automated planning application. METHODS: A software application integrated through the treatment planning system application programing interface, QuickPlan, was developed to facilitate automated planning using configurable templates for contouring, knowledge-based planning structure matching, field design, and algorithm settings. Validations are performed at various levels of the planning procedure and assist in the evaluation of readiness of the CT image, structure set, and plan layout for automated planning. QuickPlan is evaluated dosimetrically against 22 hippocampal-avoidance whole brain radiotherapy patients. The required times to treatment plan generation are compared for the validations set as well as 10 prospective patients whose plans have been automated by QuickPlan. RESULTS: The generations of 22 automated treatment plans are compared against a manual replanning using an identical process, resulting in dosimetric differences of minor clinical significance. The target dose to 2% volume and homogeneity index result in significantly decreased values for automated plans, whereas other dose metric evaluations are nonsignificant. The time to generate the treatment plans is reduced for all automated plans with a median difference of 9' 50″ ± 4' 33″. CONCLUSIONS: Template-based automated planning allows for reduced treatment planning time with consistent optimization structure creation, treatment field creation, plan optimization, and dose calculation with similar dosimetric quality. This process has potential expansion to numerous disease sites.

Phase I trial of ATM inhibitor M3541 in combination with palliative radiotherapy in patients with solid tumors.

BACKGROUND: Ataxia telangiectasia mutated (ATM) kinase orchestrates DNA double strand break (DSB) repair; ATM inhibitors may therefore enhance the therapeutic effect of DSB-inducing treatments such as radiotherapy (RT). M3541 is an orally administered selective inhibitor of ATM. METHODS: This phase I dose-escalation study evaluated the maximum-tolerated dose (MTD), recommended phase II dose(s) (RP2D), safety, pharmacokinetics (PK) and antitumor activity of M3541 in combination with fractionated palliative RT in patients with solid tumors. Fifteen patients received palliative RT (30 Gy in 10 fractions) and escalating doses of M3541 (50-300 mg administered on RT fraction days) guided by a Bayesian 2-parameter logistic regression model with overdose control. RESULTS: Doses of M3541 up to 300 mg/fraction day were well tolerated. One patient (200 mg group) experienced two dose-limiting toxicities (urinary tract infection, febrile neutropenia) that resolved with antibiotics. All patients reported ≥ 1 treatment-emergent adverse event (TEAE) but none led to treatment discontinuation. No grade ≥ 4 TEAEs were reported and there was no indication of a dose effect for any TEAE. Three patients (20.0%; 95% confidence interval 4.3-48.1) had confirmed complete or partial response. M3541 total plasma levels did not increase with dose following single or repeated dosing. No relationship was observed between dose and changes in the ratio of phosphorylated to total ATM or in immune cell counts. CONCLUSIONS: The MTD and RP2D could not be established as the study closed early due to the absence of a dose-response relationship and non-optimal PK profile. No further clinical development of M3541 was pursued. (Trial registration number ClinicalTrials.gov NCT03225105. Registration date July 21, 2017).

Phase I/II Trial of Electrophysiology-Guided Noninvasive Cardiac Radioablation for Ventricular Tachycardia.

BACKGROUND: Case studies have suggested the efficacy of catheter-free, electrophysiology-guided noninvasive cardiac radioablation for ventricular tachycardia (VT) using stereotactic body radiation therapy, although prospective data are lacking. METHODS: We conducted a prospective phase I/II trial of noninvasive cardiac radioablation in adults with treatment-refractory episodes of VT or cardiomyopathy related to premature ventricular contractions (PVCs). Arrhythmogenic scar regions were targeted by combining noninvasive anatomic and electric cardiac imaging with a standard stereotactic body radiation therapy workflow followed by delivery of a single fraction of 25 Gy to the target. The primary safety end point was treatment-related serious adverse events in the first 90 days. The primary efficacy end point was any reduction in VT episodes (tracked by indwelling implantable cardioverter defibrillators) or any reduction in PVC burden (as measured by a 24-hour Holter monitor) comparing the 6 months before and after treatment (with a 6-week blanking window after treatment). Health-related quality of life was assessed using the Short Form-36 questionnaire. RESULTS: Nineteen patients were enrolled (17 for VT, 2 for PVC cardiomyopathy). Median noninvasive ablation time was 15.3 minutes (range, 5.4-32.3). In the first 90 days, 2/19 patients (10.5%) developed a treatment-related serious adverse event. The median number of VT episodes was reduced from 119 (range, 4-292) to 3 (range, 0-31; P<0.001). Reduction was observed for both implantable cardioverter defibrillator shocks and antitachycardia pacing. VT episodes or PVC burden were reduced in 17/18 evaluable patients (94%). The frequency of VT episodes or PVC burden was reduced by 75% in 89% of patients. Overall survival was 89% at 6 months and 72% at 12 months. Use of dual antiarrhythmic medications decreased from 59% to 12% ( P=0.008). Quality of life improved in 5 of 9 Short Form-36 domains at 6 months. CONCLUSIONS: Noninvasive electrophysiology-guided cardiac radioablation is associated with markedly reduced ventricular arrhythmia burden with modest short-term risks, reduction in antiarrhythmic drug use, and improvement in quality of life. CLINICAL TRIAL REGISTRATION: URL: https://www.clinicaltrials.gov/ . Unique identifier: NCT02919618.

Noninvasive Cardiac Radiation for Ablation of Ventricular Tachycardia.

BACKGROUND: Recent advances have enabled noninvasive mapping of cardiac arrhythmias with electrocardiographic imaging and noninvasive delivery of precise ablative radiation with stereotactic body radiation therapy (SBRT). We combined these techniques to perform catheter-free, electrophysiology-guided, noninvasive cardiac radioablation for ventricular tachycardia. METHODS: We targeted arrhythmogenic scar regions by combining anatomical imaging with noninvasive electrocardiographic imaging during ventricular tachycardia that was induced by means of an implantable cardioverter-defibrillator (ICD). SBRT simulation, planning, and treatments were performed with the use of standard techniques. Patients were treated with a single fraction of 25 Gy while awake. Efficacy was assessed by counting episodes of ventricular tachycardia, as recorded by ICDs. Safety was assessed by means of serial cardiac and thoracic imaging. RESULTS: From April through November 2015, five patients with high-risk, refractory ventricular tachycardia underwent treatment. The mean noninvasive ablation time was 14 minutes (range, 11 to 18). During the 3 months before treatment, the patients had a combined history of 6577 episodes of ventricular tachycardia. During a 6-week postablation "blanking period" (when arrhythmias may occur owing to postablation inflammation), there were 680 episodes of ventricular tachycardia. After the 6-week blanking period, there were 4 episodes of ventricular tachycardia over the next 46 patient-months, for a reduction from baseline of 99.9%. A reduction in episodes of ventricular tachycardia occurred in all five patients. The mean left ventricular ejection fraction did not decrease with treatment. At 3 months, adjacent lung showed opacities consistent with mild inflammatory changes, which had resolved by 1 year. CONCLUSIONS: In five patients with refractory ventricular tachycardia, noninvasive treatment with electrophysiology-guided cardiac radioablation markedly reduced the burden of ventricular tachycardia. (Funded by Barnes-Jewish Hospital Foundation and others.).

Immunotherapy and Radiation Therapy for Non-Small Cell Lung Cancer-A Stimulating Partnership.

Non-small cell lung cancer (NSCLC) is the most common subtype of lung cancer and the leading cause of cancer-related death. Although durable local control rates are high after surgical resection or definitive radiotherapy for early-stage disease, a substantial proportion of these patients eventually experience regional and/or distant failure and succumb to their metastatic disease. The discovery of immunotherapeutics and targeted biologics has revolutionized the treatment of locally advanced and metastatic disease, improving progression-free and overall survival when incorporated with the current standards of care. Notably, post-hoc analyses and early clinical trials provide a growing body of evidence to support a synergistic effect between radiation and immunotherapy for the treatment of NSCLC from early-stage to metastatic disease. Radiotherapy appears to be capable of not only potentiating the effect of immunotherapy in targeted lesions, but also eliciting an antitumor response in distant lesions without any direct exposure to radiation. This review explores the biologic basis of immunotherapy, targeted biologics, and radiotherapy as well as the preclinical and clinical data that support the combined use of radioimmunotherapy for early-stage, locally advanced, and metastatic NSCLC.

Characterization and validation of an intra-fraction motion management system for masked-based radiosurgery.

PURPOSE: Characterize the intra-fraction motion management (IFMM) system found on the Gamma Knife Icon (GKI), including spatial accuracy, latency, temporal performance, and overall effect on delivered dose. METHODS: A phantom was constructed, consisting of a three-axis translation mount, a remote motorized flipper, and a thermoplastic sphere surrounding a radiation detector. An infrared marker was placed on the translation mount secured to the flipper. The spatial accuracy of the IFMM was measured via the translation mount in all Cartesian planes. The detector was centered at the radiation focal point. A remote signal was used to move the marker out of the IFMM tolerance and pause the beam. A two-channel electrometer was used to record the signals from the detector and the flipper when motion was signaled. These signals determined the latency and temporal performance of the GKI. RESULTS: The spatial accuracy of the IFMM was found to be <0.1 mm. The measured latency was <200 ms. The dose difference with five interruptions was <0.5%. CONCLUSION: This work provides a quantitative characterization of the GKI IFMM system as required by the Nuclear Regulatory Commission. This provides a methodology for GKI users to satisfy these requirements using common laboratory equipment in lieu of a commercial solution.

Anthropogenic disturbance homogenizes seagrass fish communities.

Anthropogenic activities have led to the biotic homogenization of many ecological communities, yet in coastal systems this phenomenon remains understudied. In particular, activities that locally affect marine habitat-forming foundation species may perturb habitat and promote species with generalist, opportunistic traits, in turn affecting spatial patterns of biodiversity. Here, we quantified fish diversity in seagrass communities across 89 sites spanning 6° latitude along the Pacific coast of Canada, to test the hypothesis that anthropogenic disturbances homogenize (i.e., lower beta-diversity) assemblages within coastal ecosystems. We test for patterns of biotic homogenization at sites within different anthropogenic disturbance categories (low, medium, and high) at two spatial scales (within and across regions) using both abundance- and incidence-based beta-diversity metrics. Our models provide clear evidence that fish communities in high anthropogenic disturbance seagrass areas are homogenized relative to those in low disturbance areas. These results were consistent across within-region comparisons using abundance- and incidence-based measures of beta-diversity, and in across-region comparisons using incidence-based measures. Physical and biotic characteristics of seagrass meadows also influenced fish beta-diversity. Biotic habitat characteristics including seagrass biomass and shoot density were more differentiated among high disturbance sites, potentially indicative of a perturbed environment. Indicator species and trait analyses revealed fishes associated with low disturbance sites had characteristics including stenotopy, lower swimming ability, and egg guarding behavior. Our study is the first to show biotic homogenization of fishes across seagrass meadows within areas of relatively high human impact. These results support the importance of targeting conservation efforts in low anthropogenic disturbance areas across land- and seascapes, as well as managing anthropogenic impacts in high activity areas.

Impact of concurrent chemotherapy with radiation therapy for elderly patients with newly diagnosed glioblastoma: a review of the National Cancer Data Base.

To investigate the utilization and overall survival (OS) impact of concurrent chemotherapy in combination with radiation therapy (RT) for elderly glioblastoma (GBM) patients. Elderly patients (age >70) with supratentorial and nonmetastatic GBM who received RT of 20-75 Gy with concurrent single-agent chemotherapy (ChemoRT) or without (RT alone) during 2004-2012 were identified from the National Cancer Data Base (NCDB). The Cochran-Armitage test was used for trend analysis. Hazard ratios (HR) and 95% confidence intervals (CIs) were determined using Cox proportional hazards. Propensity score analysis was performed to reduce selection bias in treatment allocation. A total of 5252 patients were identified (RT alone: n = 1389; ChemoRT: n = 3863). There was increasing utilization of chemotherapy during this period (45-80%, P < .001). A similar trend was also observed for the subset of age >80 (25-68%, P < .001). ChemoRT was associated with significantly better OS than RT alone (HR 0.79, 95% CI 0.70-0.89, P < .001) on multivariate analysis, and similar OS benefit was demonstrated with 1202 pairs of propensity-matched patients (HR 0.79, 95% CI 0.73-0.86, P < .001). For the matched pair, the median OS was 5.8 months with ChemoRT and 5.0 months with RT alone; the 2-year OS rate was 9% with ChemoRT and 4% with RT alone (P < .001). Concurrent chemotherapy has been administered with RT for the majority of elderly GBM patients. Addition of chemotherapy to RT for elderly GBM patients is associated with significantly improve OS in routine clinical practice.

Intensity modulated radiation therapy for recurrent ovarian cancer refractory to chemotherapy.

OBJECTIVE: To evaluate local control, survival outcomes, and toxicity after intensity modulated radiotherapy (IMRT) for recurrent chemorefractory ovarian cancer. METHODS: Between 2006 and 2014, 33 patients were treated with IMRT for recurrent ovarian cancer. Patients received a median of 3 chemotherapy regimens prior to IMRT (range, 1-12) with 11 (33%) undergoing concurrent therapy. Local control (LC), recurrence free survival (RFS), and overall survival (OS) were calculated via Kaplan-Meier method. Toxicity was assessed using the Common Terminology Criteria for Adverse Events (CTCAE) v4.03. Impact of patient characteristics on outcomes was evaluated via Cox's proportional hazard model. RESULTS: Median follow up was 23.7 months. Forty-nine sites were treated to a median dose of 5040cGy (range, 4500-7000). Nine (18%) of the 49 sites had in-field failures. Two year actuarial LC, RFS, and OS were 82%, 11%, and 63%, respectively. Seventeen patients had both a pre and post-treatment FDG-PET/CT; 6 (35%) had a complete metabolic response while 11 (65%) had a partial metabolic response. Acute ≥ grade 3 gastrointestinal (GI) toxicities occurred in 2 (6%) patients, late ≥ grade 3 GI toxicities occurred in 12 (36%), acute ≥ grade 3 hematological toxicities occurred in 5 (15%) and late ≥ grade 3 hematological toxicities occurred in 14 (42%). CONCLUSIONS: IMRT for recurrent chemorefractory ovarian cancer is associated with excellent local control and limited radiation related toxicity. Future studies will be required to determine which subpopulation will benefit most from IMRT and whether alternative techniques such as stereotactic body radiotherapy may be feasible.

The Metastatic Spine Disease Multidisciplinary Working Group Algorithms.

The Metastatic Spine Disease Multidisciplinary Working Group consists of medical and radiation oncologists, surgeons, and interventional radiologists from multiple comprehensive cancer centers who have developed evidence- and expert opinion-based algorithms for managing metastatic spine disease. The purpose of these algorithms is to facilitate interdisciplinary referrals by providing physicians with straightforward recommendations regarding the use of available treatment options, including emerging modalities such as stereotactic body radiation therapy and percutaneous tumor ablation. This consensus document details the evidence supporting the Working Group algorithms and includes illustrative cases to demonstrate how the algorithms may be applied.

Small metal artifact detection and inpainting in cardiac CT images.

Background: Quantification of cardiac motion on pre-treatment CT imaging for stereotactic arrhythmia radiotherapy patients is difficult due to the presence of image artifacts caused by metal leads of implantable cardioverter-defibrillators (ICDs). The CT scanners' onboard metal artifact reduction tool does not sufficiently reduce these artifacts. More advanced artifact reduction techniques require the raw CT projection data and thus are not applicable to already reconstructed CT images. New methods are needed to accurately reduce the metal artifacts in already reconstructed CTs to recover the otherwise lost anatomical information. Purpose: To develop a methodology to automatically detect metal artifacts in cardiac CT scans and inpaint the affected volume with anatomically consistent structures and values. Methods: Breath-hold ECG-gated 4DCT scans of 12 patients who underwent cardiac radiation therapy for treating ventricular tachycardia were collected. The metal artifacts in the images caused by the ICD leads were manually contoured. A 2D U-Net deep learning (DL) model was developed to segment the metal artifacts automatically using eight patients for training, two for validation, and two for testing. A dataset of 592 synthetic CTs was prepared by adding segmented metal artifacts from the patient 4DCT images to artifact-free cardiac CTs of 148 patients. A 3D image inpainting DL model was trained to refill the metal artifact portion in the synthetic images with realistic image contents that approached the ground truth artifact-free images. The trained inpainting model was evaluated by analyzing the automated segmentation results of the four heart chambers with and without artifacts on the synthetic dataset. Additionally, the raw cardiac patient images with metal artifacts were processed using the inpainting model and the results of metal artifact reduction were qualitatively inspected. Results: The artifact detection model worked well and produced a Dice score of 0.958 ± 0.008. The inpainting model for synthesized cases was able to recreate images that were nearly identical to the ground truth with a structural similarity index of 0.988 ± 0.012. With the chamber segmentations on the artifact-free images as the reference, the average surface Dice scores improved from 0.684 ± 0.247 to 0.964 ± 0.067 and the Hausdorff distance reduced from 3.4 ± 3.9 mm to 0.7 ± 0.7 mm. The inpainting model's use on cardiac patient CTs was visually inspected and the artifact-inpainted images were visually plausible. Conclusion: We successfully developed two deep models to detect and inpaint metal artifacts in cardiac CT images. These deep models are useful to improve the heart chamber segmentation and cardiac motion analysis in CT images corrupted by mental artifacts. The trained models and example data are available to the public through GitHub.

The First Reported Case of Treating the Ultra-Central Thorax With Cone Beam Computed Tomography-Guided Stereotactic Adaptive Radiotherapy (CT-STAR).

Stereotactic body radiotherapy (SBRT) to the central and ultra-central thorax is associated with infrequent but potentially serious adverse events. Adaptive SBRT, which provides more precise treatment planning and inter-fraction motion management, may allow the delivery of ablative doses to ultra-central tumors with effective local control and improved toxicity profiles. Herein, we describe the first reported case of cone beam computed tomography (CBCT)-guided stereotactic adaptive radiotherapy (CT-STAR) in the treatment of ultra-central non-small cell lung cancer (NSCLC) in a prospective clinical trial (NCT05785845). An 80-year-old man with radiographically diagnosed early-stage NSCLC presented for definitive management of an enlarging ultra-central lung nodule. He was prescribed 55 Gy in five fractions with CT-STAR. A simulation was performed using four-dimensional CT, and patients were planned for treatment at end-exhale breath-hold. Treatment plans were generated using a strict isotoxicity approach, which prioritized organ at risk (OAR) constraints over target coverage. During treatment, daily CBCTs were acquired and used to generate adapted contours and treatment plans based on the patient's anatomy-of-the-day, all while the patient was on the treatment table. The initial and adapted plans were compared using dose-volume histograms, and the superior plan was selected for treatment. The adapted plan was deemed superior and used for treatment in three out of five fractions. The adapted plan provided improved target coverage in two fractions and resolved an OAR hard constraint violation in one fraction. We report the successful treatment of a patient with ultra-central NSCLC utilizing CT-STAR. This case report builds on previously published in silico data to support the viability and dosimetric advantages of CT-STAR in the ablative treatment of this challenging tumor location. Further data are needed to confirm the toxicity and efficacy of this technique.

Inter-fractional portability of deep learning models for lung target tracking on cine imaging acquired in MRI-guided radiotherapy.

MRI-guided radiotherapy systems enable beam gating by tracking the target on planar, two-dimensional cine images acquired during treatment. This study aims to evaluate how deep-learning (DL) models for target tracking that are trained on data from one fraction can be translated to subsequent fractions. Cine images were acquired for six patients treated on an MRI-guided radiotherapy platform (MRIdian, Viewray Inc.) with an onboard 0.35 T MRI scanner. Three DL models (U-net, attention U-net and nested U-net) for target tracking were trained using two training strategies: (1) uniform training using data obtained only from the first fraction with testing performed on data from subsequent fractions and (2) adaptive training in which training was updated each fraction by adding 20 samples from the current fraction with testing performed on the remaining images from that fraction. Tracking performance was compared between algorithms, models and training strategies by evaluating the Dice similarity coefficient (DSC) and 95% Hausdorff Distance (HD95) between automatically generated and manually specified contours. The mean DSC for all six patients in comparing manual contours and contours generated by the onboard algorithm (OBT) were 0.68 ± 0.16. Compared to OBT, the DSC values improved 17.0 - 19.3% for the three DL models with uniform training, and 24.7 - 25.7% for the models based on adaptive training. The HD95 values improved 50.6 - 54.5% for the models based on adaptive training. DL-based techniques achieved better tracking performance than the onboard, registration-based tracking approach. DL-based tracking performance improved when implementing an adaptive strategy that augments training data fraction-by-fraction.

Using Integrin αvß6-Targeted Positron Emission Tomography Imaging to Longitudinally Monitor Radiation-Induced Pulmonary Fibrosis In Vivo.

PURPOSE: Radiation-induced pulmonary fibrosis (RIPF) is a potentially serious and disabling late complication of radiation therapy. Monitoring RIPF progression is challenging due to the absence of early detection tools and the difficulty in distinguishing RIPF from other lung diseases using standard imaging methods. In the lungs, integrin αvß6 is crucial in the development of RIPF, acting as a significant activator of transforming growth factor ß after radiation injury. This study aimed to investigate integrin αvß6-targeted positron emission tomography (PET) imaging ([64Cu]Cu-αvß6-BP) to study RIPF development in vivo. METHODS AND MATERIALS: We used a focal RIPF model (70 Gy delivered focally to a 3 mm spot in the lung) and a whole lung RIPF model (14 Gy delivered to the whole lung) in adult C57BL/6J mice. Small animal PET/computed tomography images were acquired 1 hour postinjection of 11.1 MBq of [64Cu]Cu-αvß6-BP. Animals were imaged for 8 weeks in the focal RIPF model and 6 months in the whole lung RIPF model. Immunohistochemistry for integrin αvß6 and trichrome staining were performed. RESULTS: In the focal RIPF model, there was focal uptake of [64Cu]Cu-αvß6-BP in the irradiated region at week 4 that progressively increased at weeks 6 and 8. In the whole lung RIPF model, minimal uptake of the probe was observed at 4 months post-radiation therapy, which significantly increased at months 5 and 6. Expression of integrin αvß6 was validated histologically by immunohistochemistry in both models. CONCLUSIONS: Integrin αvß6-targeted PET imaging using [64Cu]Cu-αvß6-BP can serve as a useful tool to identify RIPF in vivo.