Preliminary Evaluation of PTV Margins for Online Adaptive Radiation Therapy of the Prostatic Fossa.

PURPOSE: In modern trials, traditional planning target volume (PTV) margins for postoperative prostate radiation therapy have been large (7-10 mm) to account for both daily changes in patient positioning and target deformation. With daily adaptive radiation therapy, these interfractional changes could be minimized, potentially reducing the margins required for treatment and improving adjacent normal-tissue dosimetry. METHODS AND MATERIALS: A single-center retrospective study was conducted from March 2021 to November 2021. Patients receiving conventionally fractionated postoperative radiation therapy (PORT) for prostate cancer with pretreatment and posttreatment cone beam computed tomography (CBCT) imaging (pre-CBCT and post-CBCT, respectively) were included (248 paired images). Pretreatment and posttreatment clinical target volumes (pre-CTVs and post-CTVs) were contoured by a single observer on all CBCTs and verified by a second observer. Motion was calculated from pre-CTV to that of the post-CTV, and predicted margins were calculated with van Herk's formula. Adequate coverage of the proposed planning target volume (PTV) margin expansions (pre-PTV) were verified by determining overlap with post-CTV. In a smaller cohort (25 paired images), dosimetric changes with the proposed online adaptive margins were compared with conventional plans in the Ethos emulator environment. RESULTS: The estimated margins predicted to achieve ≥95% CTV coverage for 90% of the population were 1.6 mm, 2.0 mm, and 2.2 mm (x-, y-, and z -xes, respectively), with 95% of the absolute region of interest displacement being within 1.9 mm, 2.8 mm, and 2.1 mm. After symmetrically expanding all pre-CTVs by 3 mm, the percentage of paired images achieving ≥95% CTV coverage was 97.1%. When comparing adaptive plans (3-mm margins) with scheduled plans (7-mm margins), rectum dosimetry significantly improved, with an average relative reduction in V40Gy[cc] of 59.2% and V65Gy[cc] of 79.5% (where V40Gy and V65Gy are defined as the volumes receiving 40 Gy and 65 Gy or higher dose, respectively). CONCLUSIONS: Online daily adaptive radiation therapy could significantly decrease PTV margins for prostatic PORT and improve rectal dosimetry, with a symmetrical expansion of 3 mm achieving excellent coverage in this cohort. These results need to be validated in a larger prospective cohort.

A Proof-of-Concept Study of Artificial Intelligence-assisted Contour Editing.

Purpose: To present a concept called artificial intelligence-assisted contour editing (AIACE) and demonstrate its feasibility. Materials and Methods: The conceptual workflow of AIACE is as follows: Given an initial contour that requires clinician editing, the clinician indicates where large editing is needed, and a trained deep learning model uses this input to update the contour. This process repeats until a clinically acceptable contour is achieved. In this retrospective, proof-of-concept study, the authors demonstrated the concept on two-dimensional (2D) axial CT images from three head-and-neck cancer datasets by simulating the interaction with the AIACE model to mimic the clinical environment. The input at each iteration was one mouse click on the desired location of the contour segment. Model performance is quantified with the Dice similarity coefficient (DSC) and 95th percentile of Hausdorff distance (HD95) based on three datasets with sample sizes of 10, 28, and 20 patients. Results: The average DSCs and HD95 values of the automatically generated initial contours were 0.82 and 4.3 mm, 0.73 and 5.6 mm, and 0.67 and 11.4 mm for the three datasets, which were improved to 0.91 and 2.1 mm, 0.86 and 2.5 mm, and 0.86 and 3.3 mm, respectively, with three mouse clicks. Each deep learning-based contour update required about 20 msec. Conclusion: The authors proposed the newly developed AIACE concept, which uses deep learning models to assist clinicians in editing contours efficiently and effectively, and demonstrated its feasibility by using 2D axial CT images from three head-and-neck cancer datasets.Keywords: Segmentation, Convolutional Neural Network (CNN), CT, Deep Learning Algorithms Supplemental material is available for this article. © RSNA, 2022.

Registration-guided deep learning image segmentation for cone beam CT-based online adaptive radiotherapy.

PURPOSE: Adaptive radiotherapy (ART), especially online ART, effectively accounts for positioning errors and anatomical changes. One key component of online ART process is accurately and efficiently delineating organs at risk (OARs) and targets on online images, such as cone beam computed tomography (CBCT). Direct application of deep learning (DL)-based segmentation to CBCT images suffered from issues such as low image quality and limited available contour labels for training. To overcome these obstacles to online CBCT segmentation, we propose a registration-guided DL (RgDL) segmentation framework that integrates image registration algorithms and DL segmentation models. METHODS: The RgDL framework is composed of two components: image registration and RgDL segmentation. The image registration algorithm transforms/deforms planning contours that were subsequently used as guidance by the DL model to obtain accurate final segmentations. We had two implementations of the proposed framework-Rig-RgDL (Rig for rigid body) and Def-RgDL (Def for deformable)-with rigid body (RB) registration or deformable image registration (DIR) as the registration algorithm, respectively, and U-Net as the DL model architecture. The two implementations of RgDL framework were trained and evaluated on seven OARs in an institutional clinical head-and-neck dataset. RESULTS: Compared to the baseline approaches using the registration or the DL alone, RgDLs achieved more accurate segmentation, as measured by higher mean Dice similarity coefficients (DSCs) and other distance-based metrics. Rig-RgDL achieved a DSC of 84.5% on seven OARs on average, higher than RB or DL alone by 4.5% and 4.7%. The average DSC of Def-RgDL was 86.5%, higher than DIR or DL alone by 2.4% and 6.7%. The inference time required by the DL model component to generate final segmentations of seven OARs was less than 1 s in RgDL. By examining the contours from RgDLs and DL case by case, we found that RgDL was less susceptible to image artifacts. We also studied how the performances of RgDL and DL vary with the size of the training dataset. The DSC of DL dropped by 12.1% as the number of training data decreased from 22 to 5, whereas RgDL only dropped by 3.4%. CONCLUSION: By incorporating the patient-specific registration guidance to a population-based DL segmentation model, RgDL framework overcame the obstacles associated with online CBCT segmentation, including low image quality and insufficient training data, and achieved better segmentation accuracy than baseline methods. The resulting segmentation accuracy and efficiency show promise for applying this RgDL framework for online ART.

Preliminary Results of Multi-Institutional Phase 1 Dose Escalation Trial Using Single-Fraction Stereotactic Partial Breast Irradiation for Early Stage Breast Cancer.

PURPOSE: We report on our early experience of our prospective multicenter phase 1 dose- escalation study of single-fraction stereotactic partial breast irradiation (S-PBI) for early stage breast cancer after partial mastectomy using a robotic stereotactic radiation system. METHODS AND MATERIALS: Thirty women with in situ or invasive breast cancer stage 0, I, or II with tumor size <3 cm treated with lumpectomy were enrolled in this phase 1 single-fraction S-PBI dose-escalation trial. Women received either 22.5, 26.5, or 30 Gy in a single fraction using a robotic stereotactic radiation system. The primary outcome was to reach tumoricidal dose of 30 Gy in a single fraction to the lumpectomy cavity without exceeding the maximum tolerated dose. Secondary outcomes were to determine dose-limiting toxicity and cosmesis. Tertiary goals were ipsilateral breast recurrence rate, distant disease-free interval, recurrence-free survival, and overall survival. RESULTS: From June 2016 to January 2021, 11, 8, and 10 patients were treated to doses of 22.5, 26.5, or 30 Gy in a single fraction, respectively, with median follow-up being 47.9, 25.1, and 16.2 months. No patients experienced acute (<90 days) grade 3 or higher treatment-related toxicity, and maximum tolerated dose was not reached. There were 2 delayed grade 3 toxicities. Four patients (13.8%) developed fat necrosis across all 3 cohorts, which compares favorably with results from other PBI trials. No dose cohort had a statistically significant cosmetic detriment from baseline to 12 months or 24 months follow-up by patient- or physician-reported global cosmetic scores. There were no reports of disease recurrence. CONCLUSIONS: This phase 1 trial demonstrates that S-PBI can be used to safely escalate dose to 30 Gy in a single fraction with low toxicity and without detriment in cosmesis relative to baseline.

Exploratory ensemble interpretable model for predicting local failure in head and neck cancer: the additive benefit of CT and intra-treatment cone-beam computed tomography features.

BACKGROUND: Local failure (LF) following chemoradiation (CRT) for head and neck cancer is associated with poor overall survival. If machine learning techniques could stratify patients at risk of treatment failure based on baseline and intra-treatment imaging, such a model could facilitate response-adapted approaches to escalate, de-escalate, or switch therapy. METHODS: A 1:2 retrospective case control cohort of patients treated at a single institution with definitive radiotherapy for head and neck cancer who failed locally, in-field at a primary or nodal structure were included. Radiomic features were extracted from baseline CT and CBCT scans at fractions 1 and 21 (delta) of radiotherapy with PyRadiomics and were selected for by: reproducibility (intra-class correlation coefficients ≥0.95), redundancy [maximum relevance and minimum redundancy (mRMR)], and informativeness [recursive feature elimination (RFE)]. Separate models predicting LF of primaries or nodes were created using the explainable boosting machine (EBM) classifier with 5-fold cross-validation for (I) clinical only, (II) radiomic only (CT1 and delta features), and (III) fused models (clinical + radiomic). Twenty-five iterations were performed, and predicted scores were averaged with a parallel ensemble design. Receiver operating characteristic curves were compared between models with paired-samples t-tests. RESULTS: The fused ensemble model for primaries (using clinical, CT1, and delta features) achieved an AUC of 0.871 with a sensitivity of 78.3% and specificity of 90.9% at the maximum Youden J statistic. The fused ensemble model trended towards improvement when compared to the clinical only ensemble model (AUC =0.788, P=0.134) but reached significance when compared to the radiomic ensemble model (AUC =0.770, P=0.017). The fused ensemble model for nodes achieved an AUC of 0.910 with a sensitivity of 100.0% and specificity of 68.0%, which also trended towards improvement when compared to the clinical model (AUC =0.865, P=0.080). CONCLUSIONS: The fused ensemble EBM model achieved high discriminatory ability at predicting LF for head and neck cancer in independent primary and nodal structures. Although an additive benefit of delta radiomics over clinical factors could not be proven, the results trended towards improvement with the fused ensemble model, which are promising and worthy of prospective investigation in a larger cohort.

Cosmetic Outcomes of a Phase 1 Dose Escalation Study of 5-Fraction Stereotactic Partial Breast Irradiation for Early Stage Breast Cancer.

PURPOSE: Our purpose was to evaluate cosmetic changes after 5-fraction adjuvant stereotactic partial breast irradiation (S-PBI). METHODS AND MATERIALS: Seventy-five women with in situ or invasive breast cancer stage 0, I, or II, with tumor size ≤3 cm, were enrolled after lumpectomy in a phase 1 dose escalation trial of S-PBI into cohorts receiving 30, 32.5, 35, 37.5, or 40 Gy in 5 fractions. Before S-PBI, 3 to 4 gold fiducial markers were placed in the lumpectomy cavity for tracking with the Synchrony respiratory tracking system. S-PBI was delivered with a CyberKnife robotic radiosurgery system. Patients and physicians evaluated global cosmesis using the Harvard Breast Cosmesis Scale. Eight independent panelists evaluated digital photography for global cosmesis and 10 subdomains at baseline and follow-up. McNemar tests were used to evaluate change in cosmesis, graded as excellent/good or fair/poor, from baseline to year 3. Wilcoxon signed rank tests were used to evaluate change in subdomains. Cohen's kappa (κ) statistic was used to estimate interobserver agreement (IOA) between raters, and Fleiss' κ was used to estimate IOA between panelists. RESULTS: Median cosmetic follow-up was 5, 5, 5, 4, and 3 years for the 30, 32.5, 35, 37.5, and 40 Gy cohorts. Most patients reported excellent/good cosmesis at both baseline (86.3%) and year 3 (89.8%). No dose cohort had significantly worsened cosmesis by year 3 on McNemar analysis. No cosmetic subdomain had significant worsening by year 3. IOA was fair for patient-physician (κ = 0.300, P < .001), patient-panel (κ = 0.295, P < .001), physician-panel (κ = 0.256, P < .001), and individual panelists (Fleiss κ = 0.327, P < .001). CONCLUSIONS: Dose escalation of S-PBI from 30 to 40 Gy in 5 fractions for early stage breast cancer was not associated with a detectable change in cosmesis by year 3. S-PBI is a promising modality for treatment of early stage breast cancer.

Adaptive radiotherapy for head and neck cancer.

BACKGROUND: Although there have been dramatic improvements in radiotherapy for head and neck squamous cell carcinoma (HNSCC), including robust intensity modulation and daily image guidance, these advances are not able to account for inherent structural and spatial changes that may occur during treatment. Many sources have reported volume reductions in the primary target, nodal volumes, and parotid glands over treatment, which may result in unintended dosimetric changes affecting the side effect profile and even efficacy of the treatment. Adaptive radiotherapy (ART) is an exciting treatment paradigm that has been developed to directly adjust for these changes. MAIN BODY: Adaptive radiotherapy may be divided into two categories: anatomy-adapted (A-ART) and response-adapted ART (R-ART). Anatomy-adapted ART is the process of re-planning patients based on structural and spatial changes occurring over treatment, with the intent of reducing overdosage of sensitive structures such as the parotids, improving dose homogeneity, and preserving coverage of the target. In contrast, response-adapted ART is the process of re-planning patients based on response to treatment, such that the target and/or dose changes as a function of interim imaging during treatment, with the intent of dose escalating persistent disease and/or de-escalating surrounding normal tissue. The impact of R-ART on local control and toxicity outcomes is actively being investigated in several currently accruing trials. CONCLUSIONS: Anatomy-adapted ART is a promising modality to improve rates of xerostomia and coverage in individuals who experience significant volumetric changes during radiation, while R-ART is currently being studied to assess its utility in either dose escalation of radioresistant disease, or de-intensification of surrounding normal tissue following treatment response. In this paper, we will review the existing literature and recent advances regarding A-ART and R-ART.

Altered Cortico-Limbic Network Connectivity in Parkinsonian Depression: The Effect of Antidepressants.

BACKGROUND: Depression is a common comorbidity of Parkinson's disease (PD); however, the impact of antidepressant status on cortical function in parkinsonian depression is not fully understood. While studies of resting state functional MRI in major depression have shown that antidepressant treatment affects cortical connectivity, data on connectivity and antidepressant status in PD is sparse. OBJECTIVE: We tested the hypothesis that cortico-limbic network (CLN) resting state connectivity is abnormal in antidepressant-treated parkinsonian depression. METHODS: Thirteen antidepressant-treated depressed PD and 47 non-depressed PD participants from the Parkinson's Progression Markers Initiative (PPMI) database were included. Data was collected using 3T Siemens TIM Trio MR scanners and analyzed using SPM and CONN functional connectivity toolbox. Volumetric analysis was also performed using BrainSuite. RESULTS: We found decreased connectivity in the antidepressant-treated depressed PD group when compared to non-depressed PD between the left frontal operculum and bilateral insula, and also reduced connectivity between right orbitofrontal cortex and left temporal fusiform structures. Increased depression scores were associated with decreased insular-frontal opercular connectivity. No ROI volumetric differences were found between groups. CONCLUSION: Given the relationship between depression scores and cortico-limbic connectivity in PD, the abnormal insular-frontal opercular hypoconnectivity in this cohort may be associated with persistent depressive symptoms or antidepressant effects.

Detecting Estrogenic Ligands in Personal Care Products using a Yeast Estrogen Screen Optimized for the Undergraduate Teaching Laboratory.

The Yeast Estrogen Screen (YES) is used to detect estrogenic ligands in environmental samples and has been broadly applied in studies of endocrine disruption. Estrogenic ligands include both natural and manmade "Environmental Estrogens" (EEs) found in many consumer goods including Personal Care Products (PCPs), plastics, pesticides, and foods. EEs disrupt hormone signaling in humans and other animals, potentially reducing fertility and increasing disease risk. Despite the importance of EEs and other Endocrine Disrupting Chemicals (EDCs) to public health, endocrine disruption is not typically included in undergraduate curricula. This shortcoming is partly due to a lack of relevant laboratory activities that illustrate the principles involved while also being accessible to undergraduate students. This article presents an optimized YES for quantifying ligands in personal care products that bind estrogen receptors alpha (ERα) and/or beta (ERß). The method incorporates one of the two colorimetric substrates (ortho-nitrophenyl-ß-D-galactopyranoside (ONPG) or chlorophenol red-ß-D-galactopyranoside (CPRG)) that are cleaved by ß-galactosidase, a 6-day refrigerated incubation step to facilitate use in undergraduate laboratory courses, an automated application for LacZ calculations, and R code for the associated 4-parameter logistic regression analysis. The protocol has been designed to allow undergraduate students to develop and conduct experiments in which they screen products of their choosing for estrogen mimics. In the process, they learn about endocrine disruption, cell culture, receptor binding, enzyme activity, genetic engineering, statistics, and experimental design. Simultaneously, they also practice fundamental and broadly applicable laboratory skills, such as: calculating concentrations; making solutions; demonstrating sterile technique; serially diluting standards; constructing and interpolating standard curves; identifying variables and controls; collecting, organizing, and analyzing data; constructing and interpreting graphs; and using common laboratory equipment such as micropipettors and spectrophotometers. Thus, implementing this assay encourages students to engage in inquiry-based learning while exploring emerging issues in environmental science and health.