Stress-testing pelvic autosegmentation algorithms using anatomical edge cases.

Commercial autosegmentation has entered clinical use, however real-world performance may suffer in certain cases. We aimed to assess the influence of anatomic variants on performance. We identified 112 prostate cancer patients with anatomic variations (edge cases). Pelvic anatomy was autosegmented using three commercial tools. To evaluate performance, Dice similarity coefficients, and mean surface and 95% Hausdorff distances were calculated versus clinician-delineated references. Deep learning autosegmentation outperformed atlas-based and model-based methods. However, edge case performance was lower versus the normal cohort (0.12 mean DSC reduction). Anatomic variation presents challenges to commercial autosegmentation.

Outcomes of carotid-sparing IMRT for T1 glottic cancer: Comparison with conventional radiation.

OBJECTIVES: We aim to report oncologic outcomes after conventional radiotherapy (ConRT) using opposed lateral beams and intensity-modulated radiation therapy (IMRT) for tumor (T)1 nodal (N)0 T1 N0 glottic squamous cell carcinoma. STUDY DESIGN: Retrospective case-control study. METHODS: We retrospectively reviewed demographic, disease, and treatment characteristics for patients treated at our institution during 2000 to 2013. RESULTS: One hundred fifty-three patients (71%) were treated using ConRT and 62 (29%) using IMRT. The median follow-up for all patients was 68 months. There was no statistically significant difference in 5-year local control between patients with T1a versus T1b disease (94% vs. 89%, respectively, P = 0.5). Three-year locoregional control for patients treated with ConRT was 94% compared to 97% with IMRT (P = 0.4). Three-year overall survival (OS) for patients treated with ConRT was 92.5% compared with 100% with IMRT (P = 0.1). Twelve of 14 patients with local recurrence underwent salvage surgery with 5-year ultimate locoregional control of 98.5% and 97.1% in the ConRT and IMRT cohorts, respectively (P = 0.7). Multivariate analysis showed age < 60 years (P < 0.0001) and pretreatment Eastern Cooperative Oncology Group performance status <2 (P = 0.0022) to be independent correlates of improved OS. Postradiation cerebrovascular events were in four patients in the ConRT cohort (3%), whereas no patients in the IMRT cohort suffered any events. CONCLUSION: Because the oncologic outcomes for patients treated with IMRT were excellent and IMRT allows for carotid sparing, we have transitioned to IMRT as our standard for most patients with T1 glottic cancer. LEVEL OF EVIDENCE: 3b Laryngoscope, 130:146-153, 2020.

Artificial Intelligence in Radiation Oncology.

The integration of artificial intelligence in the radiation oncologist's workflow has multiple applications and significant potential. From the initial patient encounter, artificial intelligence may aid in pretreatment disease outcome and toxicity prediction. It may subsequently aid in treatment planning, and enhanced dose optimization. Artificial intelligence may also optimize the quality assurance process and support a higher level of safety, quality, and efficiency of care. This article describes components of the radiation consultation, planning, and treatment process and how the thoughtful integration of artificial intelligence may improve shared decision making, planning efficiency, planning quality, patient safety, and patient outcomes.

Imaging-Genomic Study of Head and Neck Squamous Cell Carcinoma: Associations Between Radiomic Phenotypes and Genomic Mechanisms via Integration of The Cancer Genome Atlas and The Cancer Imaging Archive.

PURPOSE: Recent data suggest that imaging radiomic features of a tumor could be indicative of important genomic biomarkers. Understanding the relationship between radiomic and genomic features is important for basic cancer research and future patient care. We performed a comprehensive study to discover the imaginggenomic associations in head and neck squamous cell carcinoma (HNSCC) and explore the potential of predicting tumor genomic alternations using radiomic features. METHODS: Our retrospective study integrated whole-genome multiomics data from The Cancer Genome Atlas with matched computed tomography imaging data from The Cancer Imaging Archive for the same set of 126 patients with HNSCC. Linear regression and gene set enrichment analysis were used to identify statistically significant associations between radiomic imaging and genomic features. Random forest classifier was used to predict the status of two key HNSCC molecular biomarkers, human papillomavirus and disruptive TP53 mutation, on the basis of radiomic features. RESULTS: Widespread and statistically significant associations were discovered between genomic features (including microRNA expression, somatic mutations, and transcriptional activity, copy number variations, and promoter region DNA methylation changes of pathways) and radiomic features characterizing the size, shape, and texture of tumor. Prediction of human papillomavirus and TP53 mutation status using radiomic features achieved areas under the receiver operating characteristic curve of 0.71 and 0.641, respectively. CONCLUSION: Our exploratory study suggests that radiomic features are associated with genomic characteristics at multiple molecular layers in HNSCC and provides justification for continued development of radiomics as biomarkers for relevant genomic alterations in HNSCC.

Machine Learning Applications in Head and Neck Radiation Oncology: Lessons From Open-Source Radiomics Challenges.

Radiomics leverages existing image datasets to provide non-visible data extraction via image post-processing, with the aim of identifying prognostic, and predictive imaging features at a sub-region of interest level. However, the application of radiomics is hampered by several challenges such as lack of image acquisition/analysis method standardization, impeding generalizability. As of yet, radiomics remains intriguing, but not clinically validated. We aimed to test the feasibility of a non-custom-constructed platform for disseminating existing large, standardized databases across institutions for promoting radiomics studies. Hence, University of Texas MD Anderson Cancer Center organized two public radiomics challenges in head and neck radiation oncology domain. This was done in conjunction with MICCAI 2016 satellite symposium using Kaggle-in-Class, a machine-learning and predictive analytics platform. We drew on clinical data matched to radiomics data derived from diagnostic contrast-enhanced computed tomography (CECT) images in a dataset of 315 patients with oropharyngeal cancer. Contestants were tasked to develop models for (i) classifying patients according to their human papillomavirus status, or (ii) predicting local tumor recurrence, following radiotherapy. Data were split into training, and test sets. Seventeen teams from various professional domains participated in one or both of the challenges. This review paper was based on the contestants' feedback; provided by 8 contestants only (47%). Six contestants (75%) incorporated extracted radiomics features into their predictive model building, either alone (n = 5; 62.5%), as was the case with the winner of the "HPV" challenge, or in conjunction with matched clinical attributes (n = 2; 25%). Only 23% of contestants, notably, including the winner of the "local recurrence" challenge, built their model relying solely on clinical data. In addition to the value of the integration of machine learning into clinical decision-making, our experience sheds light on challenges in sharing and directing existing datasets toward clinical applications of radiomics, including hyper-dimensionality of the clinical/imaging data attributes. Our experience may help guide researchers to create a framework for sharing and reuse of already published data that we believe will ultimately accelerate the pace of clinical applications of radiomics; both in challenge or clinical settings.

Age-adjusted comorbidity and survival in locally advanced laryngeal cancer.

BACKGROUND: The purpose of this study was to quantify the relationship among age, pretreatment comorbidity, and survival outcomes in patients with locally advanced laryngeal cancer. METHODS: Baseline comorbidity data were collected and age-adjusted Charlson Comorbidity Index (CCI) was calculated for each case. Kaplan-Meier and Cox proportional hazards modeling were used to determine associations with survival. RESULTS: For 548 patients, with a median age of 59 years (range 31-91 years), 58% were treated with larynx preservation and the rest with total laryngectomy and adjuvant radiotherapy (RT). Two hundred thirty-eight patients (43%) had at least 1 comorbidity each. Cardiovascular diseases were the most common comorbidities (19%). The 5-year overall survival (OS) for patients with CCI ≤3 (n = 442) were superior to CCI >3 (n = 106; 60% vs 41%; P < .0001), although the 5-year disease-specific survival (DSS) rates were not significantly different. The 5-year noncancer CSS was better for age-adjusted CCI ≤3 (88% vs 67%; P < .0001). CONCLUSION: The age-adjusted CCI is a significant predictor of noncancer CSS and OS for patients with locally advanced laryngeal cancer but is not associated with DSS.

Feeding Tube Utilization in Patients with Salivary Gland Malignancies.

Objectives To evaluate feeding tube utilization in patients with salivary gland malignancies (SGMs). Study Design Case series with planned data collection. Setting The University of Texas MD Anderson Cancer Center, Houston, Texas, USA. Subjects and Methods Patients (N = 287) were sampled from an epidemiologic SGM registry during a 12-year period. Feeding tube history was retrospectively reviewed. Patients with outside locoregional therapy or palliative treatment were excluded. Enteral feeding and length of dependence were analyzed as a function treatment modality and site of SGM. Results Of 287 patients, 79 (28%) required temporary nasogastric tube feeding (median duration: 13 days, interquartile range: 6-21). Among those 79, 30 (10% of total cohort) required conversion to percutaneous gastrostomy tube (G-tube). Median G-tube duration was 4.8 months (interquartile range: 3.7-13.1). G-tube placement was necessary only in patients receiving multimodality therapy ( P < .001), and among those, 50% with SGMs arising from pharyngeal/laryngeal sites required G-tube, as compared with 8% to 19% of SGMs arising from all other sites ( P < .01). At a median follow-up of 2.4 years, 9 (3%) of all SGM patients were G-tube dependent, but 14% (3 of 22) with laryngeal/pharyngeal sites treated with multimodality therapy remained chronically G-tube dependent. Conclusion While almost 30% of SGM survivors require a temporary nasogastric tube, G-tube utilization is uncommon, in roughly 10% of SGM overall. G-tube utilization appears exclusive to patients treated with multimodality therapy, and chronic gastrostomy remains high (14%) in patients with minor gland cancers arising in the pharynx/larynx, suggesting impetus for dysphagia prophylaxis in these higher-risk subsets, similar to patients treated for squamous cancers.

Radiomics in head and neck cancer: from exploration to application.

In the context of clinical oncology, a fundamental goal of radiomics is the extraction of large amounts of quantitative features whose subsequent analysis can be used for decision support towards personalized and actionable cancer care. Head and neck cancers present a unique set of diagnostic and therapeutic challenges by nature of its complex anatomy and heterogeneity. Radiomics holds the potential to address these barriers, but only if as a collective field we direct future effort towards investigating specific oncologic function and oncologic outcomes, with external validation and collaborative multi-institutional efforts to begin standardizing and refining radiomic signatures. Here we present an overview of radiomic texture analysis methods as well as the software infrastructure, review the developments of radiomics in head and neck cancer applications, discuss unmet challenges, and propose key recommendations for moving the field forward.