Enhancing Precision in Cardiac Segmentation for MR-Guided Radiation Therapy through Deep Learning.

INTRODUCTION: Cardiac substructure dose metrics are more strongly linked to late cardiac morbidities than whole-heart metrics. MR-guided radiation therapy (MRgRT) enables substructure visualization during daily localization, allowing potential for enhanced cardiac sparing. We extend a publicly available state-of-the-art deep learning (DL) framework, nnU-Net, to incorporate self-distillation (nnU-Net.wSD) for substructure segmentation for MRgRT. METHODS: Eighteen (Institute A) patients who underwent thoracic or abdominal radiation therapy on a 0.35 T MR-guided linac were retrospectively evaluated. On each image, one of two radiation oncologists delineated reference contours of 12 cardiac substructures (chambers, great vessels, and coronary arteries) used to train (n=10), validate (n=3), and test (n=5) nnU-Net.wSD leveraging a teacher-student network and comparing to standard 3D U-Net. The impact of using simulation data or including 3-4 daily images for augmentation during training was evaluated for nnU-Net.wSD. Geometric metrics (Dice similarity coefficient (DSC), mean distance to agreement (MDA), and 95% Hausdorff distance (HD95)), visual inspection, and clinical dose volume histograms (DVHs) were evaluated. To determine generalizability, Institute A's model was tested on an unlabeled dataset from Institute B (n=22) and evaluated via consensus scoring and volume comparisons. RESULTS: nnU-Net.wSD yielded a DSC (reported mean ± standard deviation) of 0.65±0.25 across the 12 substructures (Chambers: 0.85±0.05, Great Vessels: 0.67±0.19, and Coronary Arteries 0.33±0.16, mean MDA <3 mm, and mean HD95 <9 mm) while outperforming the 3D U-Net (0.583±0.28, p<0.01). Leveraging fractionated data for augmentation improved over a single MR-SIM timepoint (0.579±0.29, p<0.01). Predicted contours yielded DVHs that closely matched the clinical treatment plans where mean and D0.03cc doses deviated by 0.32±0.5 Gy and 1.42±2.6 Gy respectively. No statistically significant differences between Institute A and B volumes (p>0.05) for 11 of 12 substructures with larger volumes requiring minor changes and coronary arteries exhibiting more variability. CONCLUSIONS: This work is a critical step to rapid and reliable cardiac substructure segmentation to improve cardiac sparing in low-field MRgRT.

Recurrence Risk Stratification for Women With FIGO Stage I Uterine Endometrioid Carcinoma Who Underwent Surgical Lymph Node Evaluation.

OBJECTIVE: The aim of this study was to estimate the recurrence risk based on the number of prognostic factors for patients with stage I uterine endometrioid carcinoma (EC) who underwent surgical lymph node evaluation (SLNE) and were managed with observation. METHODS: We queried our database for women with FIGO-2009 stage I EC who underwent surgical staging including SLNE. Multivariate analysis with stepwise model selection was used to determine independent risk factors for 5-year recurrence-free survival (RFS). Study groups based on risk factors were compared for RFS, disease-specific survival, and overall survival. RESULTS: A total of 706 patients were identified: median age was 60 years (range, 30 to 93 y) and median follow-up was 120 months. Median number of examined lymph nodes was 8 (range, 1 to 66). 91% were stage IA, 75% had grade 1 and lymphovascular space invasion was detected in 6%. Independent predictors of 5-year RFS included age 60 years and above ( P =0.038), grade 2 ( P =0.003), and grade 3 ( P <0.001) versus grade 1. Five-year RFS for group 0 (age less than 60 y and grade 1) was 98% versus 92% for group 1 (either: age 60 y and older or grade 2/3) versus 84% for group 2 (both: age 60 y and above and grade 2/3), respectively ( P <0.001). Five-year disease-specific survival was 100% versus 98% versus 95%, ( P =0.012) and 5-year overall survival was 98% versus 90% versus 81%, for groups 0, 1, and 2, respectively ( P <0.001). CONCLUSIONS: In patients with stage I EC who received SLNE and no adjuvant therapy, only age 60 years and above and high tumor grade were independent predictors of recurrence and can be used to quantify individualized recurrence risk, whereas lymphovascular space invasion was not an independent prognostic factor in this cohort.

The Prognostic Significance of the Depth of Cervical Stromal Invasion in Women With FIGO Stage II Uterine Endometrioid Carcinoma.

OBJECTIVE: The objective of this study was to investigate the prognostic significance of the depth of cervical stromal invasion (CSI) in women with FIGO stage II uterine endometrioid adenocarcinoma (EC). METHODS: Our database of women with EC was quired for patients with stage II EC. Pathologic slides were retrieved and reviewed by gynecologic pathologists to determine cervical stromal thickness and depth of CSI as a percentage of stromal thickness (%CSI). Kaplan-Meier, univariate, and multivariate analyses were used to compare recurrence-free, disease-specific (DSS), and overall survival (OS) between women who had<50% versus ≥50% CSI. Univariate and multivariate analyses were used to assess other prognostic variables associated with survival endpoints. RESULTS: A total of 117 patients were included in our study who had hysterectomy between 1/1990 and 8/2021. Seventy-nine patients (68%) with <50% and 38 (32w%) with ≥50% CSI. After a median follow-up of 131 months, 5-year DSS was significantly worse for women with ≥50% CSI (78% vs. 91%; P =0.04). However, %CSI was not an independent predictor for any of the studied survival endpoints. Independent predictors of worse 5-year recurrence-free survival and DSS included FIGO grade 3 tumors ( P =0.02) and the presence of lymphovascular space invasion ( P =0.03). Grade 3 tumors were the only independent predictor of worse 5-year OS ( P =0.02). CONCLUSIONS: Our results suggest that deep CSI is not an independent prognostic factor for survival endpoints in women with stage II uterine endometroid adenocarcinoma. The lack of independent prognostic significance of the depth CSI needs to be validated in a multi-institutional analysis.

Cetuximab and anemia prevention in head and neck cancer patients undergoing radiotherapy.

BACKGROUND: Epidermal growth factor receptor (EGFR) activation is associated with increased production of interleukin 6 (IL6), which is intensified by radiotherapy (RT) induced inflammatory response. Elevated IL6 levels intensifies RT-induced anemia by upregulating hepcidin causing functional iron deficiency. Cetuximab, an EGFR inhibitor, has been associated with lower rates of anemia for locally advanced head and neck squamous cell carcinoma (HNSCC). We hypothesized that concomitant cetuximab could prevent RT-induced anemia. METHODS: We queried our institutional head and neck cancers database for non-metastatic HNSCC cases that received RT with concomitant cetuximab or RT-only between 2006 and 2018. Cetuximab was administered for some high-risk cases medically unfit for platinum agents per multidisciplinary team evaluation. We only included patients who had at least one complete blood count in the 4 months preceding and after RT. We compared the prevalence of anemia (defined as hemoglobin (Hb) below 12 g/dL in females and 13 g/dL in males) and mean Hb levels at baseline and after RT. Improvement of anemia/Hb (resolution of baseline anemia and/or an increase of baseline Hb ≥1 g/dL after RT), and overall survival (OS) in relation to anemia/Hb dynamics were also compared. RESULTS: A total of 171 patients were identified equally distributed between cetuximab-plus-RT and RT-only groups. The cetuximab-plus-RT group had more locally-advanced stage, oropharyngeal and high grade tumors (p < 0.001 for all). Baseline anemia/Hb were similar, however anemia after RT conclusion was higher in the cetuximab-plus-RT vs RT-only (63.5% vs. 44.2%; p = 0.017), with a mean Hb of 11.98 g/dL vs. 12.9 g/dL; p = 0.003, for both respectively. This contributed to significantly worse anemia/Hb improvement for cetuximab-plus-RT (18.8% vs. 37.2%; p = 0.007). This effect was maintained after adjusting for other factors in multivariate analysis. The prevalence of iron, vitamin-B12 and folate deficiencies; and chronic kidney disease, was non-different. Baseline anemia was associated with worse OS (p = 0.0052) for the whole study cohort. Nevertheless, improvement of anemia/Hb was only marginally associated with better OS (p = 0.068). CONCLUSIONS: In contrast to previous studies, cetuximab was not associated with lower rates of anemia after RT for nonmetastatic HNSCC patients compared to RT-alone. Dedicated prospective studies are needed to elucidate the effect of cetuximab on RT-induced anemia.

Characterizing Sensitive Cardiac Substructure Excursion Due to Respiration.

PURPOSE: Whole-heart dose metrics are not as strongly linked to late cardiac morbidities as radiation doses to individual cardiac substructures. Our aim was to characterize the excursion and dosimetric variation throughout respiration of sensitive cardiac substructures for future robust safety margin design. METHODS AND MATERIALS: Eleven patients with cancer treatments in the thorax underwent 4-phase noncontrast 4-dimensional computed tomography (4DCT) with T2-weighted magnetic resonance imaging in end-exhale. The end-exhale phase of the 4DCT was rigidly registered with the magnetic resonance imaging and refined with an assisted alignment surrounding the heart from which 13 substructures (chambers, great vessels, coronary arteries, etc) were contoured by a radiation oncologist on the 4DCT. Contours were deformed to the other respiratory phases via an intensity-based deformable registration for radiation oncologist verification. Measurements of centroid and volume were evaluated between phases. Mean and maximum dose to substructures were evaluated across respiratory phases for the breast (n = 8) and thoracic cancer (n = 3) cohorts. RESULTS: Paired t tests revealed reasonable maintenance of geometric and anatomic properties (P < .05 for 4/39 volume comparisons). Maximum displacements >5 mm were found for 24.8%, 8.5%, and 64.5% of the cases in the left-right, anterior-posterior, and superior-inferior axes, respectively. Vector displacements were largest for the inferior vena cava and the right coronary artery, with displacements up to 17.9 mm. In breast, the left anterior descending artery Dmean varied 3.03 ± 1.75 Gy (range, 0.53-5.18 Gy) throughout respiration whereas lung showed patient-specific results. Across all patients, whole heart metrics were insensitive to breathing phase (mean and maximum dose variations <0.5 Gy). CONCLUSIONS: This study characterized the intrafraction displacement of the cardiac substructures through the respiratory cycle and highlighted their increased dosimetric sensitivity to local dose changes not captured by whole heart metrics. Results suggest value of cardiac substructure margin generation to enable more robust cardiac sparing and to reduce the effect of respiration on overall treatment plan quality.

Radiomics outperforms clinical factors in characterizing human papilloma virus (HPV) for patients with oropharyngeal squamous cell carcinomas.

Purpose.To utilize radiomic features extracted from CT images to characterize Human Papilloma Virus (HPV) for patients with oropharyngeal cancer squamous cell carcinoma (OPSCC).Methods.One hundred twenty-eight OPSCC patients with known HPV-status (60-HPV + and 68-HPV-, confirmed by immunohistochemistry-P16-protein testing) were retrospectively studied. Radiomic features (11 feature-categories) were extracted in 3D from contrast-enhanced (CE)-CT images of gross-tumor-volumes using 'in-house' software ('ROdiomiX') developed and validated following the image-biomarker-standardization-initiative (IBSI) guidelines. Six clinical factors were investigated: Age-at-Diagnosis, Gender, Total-Charlson, Alcohol-Use, Smoking-History, and T-Stage. A Least-Absolute-Shrinkage-and-Selection-Operation (Lasso) technique combined with a Generalized-Linear-Model (Lasso-GLM) were applied to perform regularization in the radiomic and clinical feature spaces to identify the ranking of optimal feature subsets with most representative information for prediction of HPV. Lasso-GLM models/classifiers based on clinical factors only, radiomics only, and combined clinical and radiomics (ensemble/integrated) were constructed using random-permutation-sampling. Tests of significance (One-way ANOVA), average Area-Under-Receiver-Operating-Characteristic (AUC), and Positive and Negative Predictive values (PPV and NPV) were computed to estimate the generalization-error and prediction performance of the classifiers.Results.Five clinical factors, including T-stage, smoking status, and age, and 14 radiomic features, including tumor morphology, and intensity contrast were found to be statistically significant discriminators between HPV positive and negative cohorts. Performances for prediction of HPV for the 3 classifiers were: Radiomics-Lasso-GLM: AUC/PPV/NPV = 0.789/0.755/0.805; Clinical-Lasso-GLM: 0.676/0.747/0.672, and Integrated/Ensemble-Lasso-GLM: 0.895/0.874/0.844. Results imply that the radiomics-based classifier enabled better characterization and performance prediction of HPV relative to clinical factors, and that the combination of both radiomics and clinical factors yields even higher accuracy characterization and predictive performance.Conclusion.Albeit subject to confirmation in a larger cohort, this pilot study presents encouraging results in support of the role of radiomic features towards characterization of HPV in patients with OPSCC.

Outcomes of Post-Operative Treatment with Concurrent Chemoradiotherapy (CRT) in High-Risk Resected Oral Cavity Squamous Cell Carcinoma (OCSCC): A Multi-Institutional Collaboration.

Adjuvant chemoradiation (CRT), with high-dose cisplatin remains standard treatment for oral cavity squamous cell carcinoma (OCSCC) with high-risk pathologic features. We evaluated outcomes associated with different cisplatin dosing and schedules, concurrent with radiation (RT), and the effect of cumulative dosing of cisplatin. An IRB-approved collaborative database of patients (pts) with primary OCSCC (Stage I-IVB AJCC 7th edition) treated with primary surgical resection between January 2005 and January 2015, with or without adjuvant therapy, was established from six academic institutions. Patients were categorized by cisplatin dose and schedule, and resultant groups compared for demographic data, pathologic features, and outcomes by statistical analysis to determine disease free survival (DFS) and freedom from metastatic disease (DM). From a total sample size of 1282 pts, 196 pts were identified with high-risk features who were treated with adjuvant CRT. Administration schedule of cisplatin was not significantly associated with DFS. On multivariate (MVA), DFS was significantly better in patients without perineural invasion (PNI) and in those receiving ≥200 mg/m2 cisplatin dose (p < 0.001 and 0.007). Median DFS, by cisplatin dose, was 10.5 (<200 mg/m2) vs. 20.8 months (≥200 mg/m2). Our analysis demonstrated cumulative cisplatin dose ≥200 mg/m2 was associated with improved DFS in high-risk resected OCSCC pts.

Quantifying inter-fraction cardiac substructure displacement during radiotherapy via magnetic resonance imaging guidance.

PURPOSE: Emerging evidence suggests cardiac substructures are highly radiosensitive during radiation therapy for cancer treatment. However, variability in substructure position after tumor localization has not been well characterized. This study quantifies inter-fraction displacement and planning organ at risk volumes (PRVs) of substructures by leveraging the excellent soft tissue contrast of magnetic resonance imaging (MRI). METHODS: Eighteen retrospectively evaluated patients underwent radiotherapy for intrathoracic tumors with a 0.35 T MRI-guided linear accelerator. Imaging was acquired at a 17-25 s breath-hold (resolution 1.5 × 1.5 × 3 mm3). Three to four daily MRIs per patient (n = 71) were rigidly registered to the planning MRI-simulation based on tumor matching. Deep learning or atlas-based segmentation propagated 13 substructures (e.g., chambers, coronary arteries, great vessels) to daily MRIs and were verified by two radiation oncologists. Daily centroid displacements from MRI-simulation were quantified and PRVs were calculated. RESULTS: Across substructures, inter-fraction displacements for 14% in the left-right, 18% in the anterior-posterior, and 21% of fractions in the superior-inferior were > 5 mm. Due to lack of breath-hold compliance, ~4% of all structures shifted > 10 mm in any axis. For the chambers, median displacements were 1.8, 1.9, and 2.2 mm in the left-right, anterior-posterior, and superior-inferior axis, respectively. Great vessels demonstrated larger displacements (> 3 mm) in the superior-inferior axis (43% of shifts) and were only 25% (left-right) and 29% (anterior-posterior) elsewhere. PRVs from 3 to 5 mm were determined as anisotropic substructure-specific margins. CONCLUSIONS: This exploratory work derived substructure-specific safety margins to ensure highly effective cardiac sparing. Findings require validation in a larger cohort for robust margin derivation and for applications in prospective clinical trials.

Evaluating compliance with process-related quality metrics and survival in oral cavity squamous cell carcinoma: Multi-institutional oral cavity collaboration study.

BACKGROUND: Process-related measures have been proposed as quality metrics in head and neck cancer care. A recent single-institution study identified four key metrics associated with increased survival. This study sought to validate the association of these quality metrics with survival in a multi-institutional cohort. METHODS: Multicenter retrospective study of patients with oral cavity squamous cell (1/2005-1/2015). Baseline patient and disease characteristics and compliance with quality metrics was evaluated. Association between compliance with quality metrics with overall survival (OS), disease-free survival (DFS), and disease-specific survival (DSS) was evaluated using Cox proportional hazards models. RESULTS: Failure to comply with two or more of the quality metrics was associated with worse OS, DFS, and DSS. Adherence to all or all but one of the quality metrics was found to be associated with improved survival. CONCLUSIONS: Process-related quality metrics are associated with increased survival in patients with oral cavity squamous cell carcinoma in a multi-institutional cohort.

Incorporating sensitive cardiac substructure sparing into radiation therapy planning.

PURPOSE: Rising evidence suggests that cardiac substructures are highly radiosensitive. However, they are not routinely considered in treatment planning as they are not readily visualized on treatment planning CTs (TPCTs). This work integrated the soft tissue contrast provided by low-field MRIs acquired on an MR-linac via image registration to further enable cardiac substructure sparing on TPCTs. METHODS: Sixteen upper thoracic patients treated at various breathing states (7 end-exhalation, 7 end-inhalation, 2 free-breathing) on a 0.35T MR-linac were retrospectively evaluated. A hybrid MR/CT atlas and a deep learning three-dimensional (3D) U-Net propagated 13 substructures to TPCTs. Radiation oncologists revised contours using registered MRIs. Clinical treatment plans were re-optimized and evaluated for beam arrangement modifications to reduce substructure doses. Dosimetric assessment included mean and maximum (0.03cc) dose, left ventricular volume receiving 5Gy (LV-V5), and other clinical endpoints. As metrics of plan complexity, total MU and treatment time were evaluated between approaches. RESULTS: Cardiac sparing plans reduced the mean heart dose (mean reduction 0.7 ± 0.6, range 0.1 to 2.5 Gy). Re-optimized plans reduced left anterior descending artery (LADA) mean and LADA0.03cc (0.0-63.9% and 0.0 to 17.3 Gy, respectively). LV0.03cc was reduced by >1.5 Gy for 10 patients while 6 cases had large reductions (>7%) in LV-V5. Left atrial mean dose was equivalent/reduced in all sparing plans (mean reduction 0.9 ± 1.2 Gy). The left main coronary artery was better spared in all cases for mean dose and D0.03cc . One patient exhibited >10 Gy reduction in D0.03cc to four substructures. There was no statistical difference in treatment time and MU, or clinical endpoints to the planning target volume, lung, esophagus, or spinal cord after re-optimization. Four patients benefited from new beam arrangements, leading to further dose reductions. CONCLUSIONS: By introducing 0.35T MRIs acquired on an MR-linac to verify cardiac substructure segmentations for CT-based treatment planning, an opportunity was presented for more effective sparing with limited increase in plan complexity. Validation in a larger cohort with appropriate margins offers potential to reduce radiation-related cardiotoxicities.

Does the Interval Between Hysterectomy and Start of Adjuvant Radiation Treatment Influence Survival in Women With Endometrial Carcinoma?: A National Cancer Database analysis.

OBJECTIVE: The objective of this study was to analyze the impact of the time interval (TI) between hysterectomy and initiation of adjuvant radiation treatment (ART) on overall survival (OS) among women with early stage endometrial carcinoma (EC) using the National Cancer Database (NCDB). MATERIALS AND METHODS: The NCDB was queried for women with the International Federation of Gynecology and Obstetrics (FIGO) stage I to II EC who underwent hysterectomy followed by ART. We examined the prognostic impact of TI on OS using the cutoff ≤8 or >8 weeks to initiate radiation treatment (RT). Two groups of patients were created. Kaplan-Meier curves were created for OS analysis. Predictors of OS were identified. RESULTS: A total 16,520 women were identified. The median follow-up time for the entire cohort was 59.1 months. Median age was 63 years, and 82% were FIGO stage I. Pelvic external beam RT alone was used in 9569 (58%) and vaginal brachytherapy alone in 4265 women (26%). In total, 10,040 women (61%) received RT ≤8 weeks. Delay in initiating RT >8 weeks was associated with shorter 5-year OS (P=0.048). Independent predictors of shorter OS includes older age, African American race, higher comorbidity burden, higher tumor grade, the presence of lymphovascular invasion and stage II tumors. Although TI in initiating RT was a significant predictor for OS in univariate analysis, its independent significance of OS was lost on multivariate analysis (P=0.28). CONCLUSION: Our study suggests that TI between hysterectomy and initiation of ART was not an independent predictor of OS in women with early stage EC.

Charlson Comorbidity score influence on prostate cancer survival and radiation-related toxicity.

INTRODUCTION: In addition to survival endpoints, we explored the impact of Charlson Comorbidity-Index (CCI) on the acute and late toxicities in men with localized prostate cancer who received dose-escalated definitive radiotherapy (RT). MATERIALS AND METHODS: CCI scores at diagnosis and survival outcomes were identified for men with intermediate/high-risk prostate cancer treated with RT (1/2007-12/2012). Study-cohort was accordingly grouped into no, mild and severe comorbidity (CCI-0, 1 or 2+). CCI-groups were compared for demographics, prognostic-factors; and RT-related toxicities based on RTOG/CTCAE criteria. Kaplan-Meier curves and Uni/multivariate (MVA) analyses were used to examine the influence of CCI-group on overall (OS), disease-specific (DSS) and biochemical-relapse free (BRFS) survival. RESULTS: We included 257 patients with median age 73 years (48-85), 53% African-American and 67% had intermediate-risk. Median prostate RT-dose was 76 Gy; and 47% received androgen-deprivation therapy. CCI-0,1,2+ groups encompassed 76 (30%), 54 (21%) and 127 (49%) patients, respectively and were well-balanced. Ten and 15-years OS were significantly different (76% versus 46% versus 55% for 10-years OS and 53% versus 31% versus 14% for 15-years OS for CCI-0 versus CCI-1[HR:2.25; CI[1.31-3.87]] versus CCI-2+[HR:2.73; CI[1.73-4.31]]; p < 0.001. CCI-0 had better DSS than CCI-2+ (HR:2.23; CI[1.06-4.68]; p = 0.03) and BRFS was similar (p = 0.99). Late G2/3 RT-toxicities were more common in CCI-2+ (47%) than CCI-1 (44%) and CCI-0 (29%), p = 0.032; with non-different acute-toxicities (p = 0.62). On MVA, increased CCI was deterministic for OS (HR:3.65; CI [1.71:7.79]; p < 0.001) and was only marginal for DSS (HR:2.55; CI [0.98-6.6]; p = 0.05) with no impact on BRFS (p > 0.05). CONCLUSIONS: Higher CCI is a significant predictor for late RT-related side-effects and shorter OS in men with localized prostate cancer. Baseline comorbidities should be considered during initial counseling and follow up visits.

Correlation of normal lung density changes with dose after stereotactic body radiotherapy (SBRT) for early stage lung cancer.

BACKGROUND AND PURPOSE: To investigate the correlation between normal lung CT density changes with dose accuracy and outcome after stereotactic body radiation therapy (SBRT) for patients with early stage non-small-cell lung cancer (NSCLC). MATERIALS AND METHODS: Thirty-one patients (with a total of 33 lesions) with non-small cell lung cancer were selected out of 270 patients treated with SBRT at a single institution between 2003 and 2009. Out of these 31 patients, 10 patients had developed radiation pneumonitis (RP). Dose distributions originally planned using a 1-D pencil beam-based dose algorithm were retrospectively recomputed using different algorithms. Prescription dose was 48 Gy in 4 fractions in most patients. Planning CT images were rigidly registered to follow-up CT datasets at 3-9 months after treatment. Corresponding dose distributions were mapped from planning to follow-up CT images. Hounsfield Unit (HU) changes in lung density in individual, 5 Gy, dose bins from 5 to 45 Gy were assessed in the peri-tumoral region. Correlations between HU changes in various normal lung regions, dose indices (V20, MLD, generalized equivalent uniform dose (gEUD)), and RP grade were investigated. RESULTS: Strong positive correlation was found between HU changes in the peri-tumoral region and RP grade (Spearman's r = 0.760; p < 0.001). Positive correlation was also observed between RP and HU changes in the region covered by V20 for all algorithms (Spearman's r ≥ 0.738; p < 0.001). Additionally, V20, MLD, and gEUD were significantly correlated with RP grade (p < 0.01). MLD in the peri-tumoral region computed with model-based algorithms was 5-7% lower than the PB-based methods. CONCLUSION: Changes of lung density in the peri-tumoral lung and in the region covered by V20 were strongly associated with RP grade. Relative to model-based methods, PB algorithms over-estimated mean peri-tumoral dose and showed displacement of the high-dose region, which correlated with HU changes on follow-up CT scans.

Application of radiomics for the prediction of HPV status for patients with head and neck cancers.

PURPOSE: To perform radiomic analysis of primary tumors extracted from pretreatment contrast-enhanced computed tomography (CE-CT) images for patients with oropharyngeal cancers to identify discriminant features and construct an optimal classifier for the characterization and prediction of human papilloma virus (HPV) status. MATERIALS AND METHODS: One hundred and eighty seven patients with oropharyngeal cancers with known HPV status (confirmed by immunohistochemistry-p16 protein testing) were retrospectively studied as follows: Group A: 95 patients (19HPV- and 76HPV+) from the MICAII grand challenge. Group B: 92 patients (52HPV- and 40HPV+) from our institution. Radiomic features (172) were extracted from pretreatment diagnostic CE-CT images of the gross tumor volume (GTV). Levene and Kolmogorov-Smirnov's tests with absolute biserial correlation (>0.48) were used to identify the discriminant features between the HPV+ and HPV- groups. The discriminant features were used to train and test eight different classifiers. Area under receiver operating characteristic (AUC), positive predictive and negative predictive values (PPV and NPV, respectively) were used to evaluate the performance of the classifiers. Principal component analysis (PCA) was applied on the discriminant feature set and seven PCs were used to train and test a generalized linear model (GLM) classifier. RESULTS: Among 172 radiomic features only 12 radiomic features (from 3 categories) were significantly different (P < 0.05, |BSC| > 0.48) between the HPV+ and HPV- groups. Among the eight classifiers trained and applied for prediction of HPV status, the GLM showed the highest performance for each discriminant feature and the combined 12 features: AUC/PPV/NPV = 0.878/0.834/0.811. The GLM high prediction power was AUC/PPV/NPV = 0.849/0.731/0.788 and AUC/PPV/NPV = 0.869/0.807/0.870 for unseen test datasets for groups A and B, respectively. After eliminating the correlation among discriminant features by applying PCA analysis, the performance of the GLM was improved by 3.3%, 2.2%, and 1.8% for AUC, PPV, and NPV, respectively. CONCLUSIONS: Results imply that GTV's for HPV+ patients exhibit higher intensities, smaller lesion size, greater sphericity/roundness, and higher spatial intensity-variation/heterogeneity. Results are suggestive that radiomic features primarily associated with the spatial arrangement and morphological appearance of the tumor on contrast-enhanced diagnostic CT datasets may be potentially used for classification of HPV status.

Cardiac substructure segmentation with deep learning for improved cardiac sparing.

PURPOSE: Radiation dose to cardiac substructures is related to radiation-induced heart disease. However, substructures are not considered in radiation therapy planning (RTP) due to poor visualization on CT. Therefore, we developed a novel deep learning (DL) pipeline leveraging MRI's soft tissue contrast coupled with CT for state-of-the-art cardiac substructure segmentation requiring a single, non-contrast CT input. MATERIALS/METHODS: Thirty-two left-sided whole-breast cancer patients underwent cardiac T2 MRI and CT-simulation. A rigid cardiac-confined MR/CT registration enabled ground truth delineations of 12 substructures (chambers, great vessels (GVs), coronary arteries (CAs), etc.). Paired MRI/CT data (25 patients) were placed into separate image channels to train a three-dimensional (3D) neural network using the entire 3D image. Deep supervision and a Dice-weighted multi-class loss function were applied. Results were assessed pre/post augmentation and post-processing (3D conditional random field (CRF)). Results for 11 test CTs (seven unique patients) were compared to ground truth and a multi-atlas method (MA) via Dice similarity coefficient (DSC), mean distance to agreement (MDA), and Wilcoxon signed-ranks tests. Three physicians evaluated clinical acceptance via consensus scoring (5-point scale). RESULTS: The model stabilized in ~19 h (200 epochs, training error <0.001). Augmentation and CRF increased DSC 5.0 ± 7.9% and 1.2 ± 2.5%, across substructures, respectively. DL provided accurate segmentations for chambers (DSC = 0.88 ± 0.03), GVs (DSC = 0.85 ± 0.03), and pulmonary veins (DSC = 0.77 ± 0.04). Combined DSC for CAs was 0.50 ± 0.14. MDA across substructures was <2.0 mm (GV MDA = 1.24 ± 0.31 mm). No substructures had statistical volume differences (P > 0.05) to ground truth. In four cases, DL yielded left main CA contours, whereas MA segmentation failed, and provided improved consensus scores in 44/60 comparisons to MA. DL provided clinically acceptable segmentations for all graded patients for 3/4 chambers. DL contour generation took ~14 s per patient. CONCLUSIONS: These promising results suggest DL poses major efficiency and accuracy gains for cardiac substructure segmentation offering high potential for rapid implementation into RTP for improved cardiac sparing.

The effect of treatment package time in head and neck cancer patients treated with adjuvant radiotherapy and concurrent systemic therapy.

OBJECTIVES: In patients with head and neck carcinoma, "treatment package time" (TPT) was proven to impact outcomes in cases receiving adjuvant radiotherapy alone. Its impact in patients receiving radiotherapy with concurrent systemic therapy has not been studied previously. The TPT influence on survival endpoints for patients treated with surgery followed by radiation and concurrent systemic therapy was analyzed. METHODS: Institutional database to identify head and neck carcinoma cases treated with definitive surgery followed by concomitant chemo(bio) radiotherapy (CRT) was used. TPT was the number of days elapsed between surgery and the last day of radiation. %FINDCUT SAS macro tool was used to search for the cutoff TPT that was associated with significant survival benefit. Kaplan-Meier curves, log-rank tests as well as univariate and multivariate analyses were used to assess overall survival (OS) and recurrence free survival (RFS). RESULTS: One hundred and three cases with a median follow up of 37 months were included in the study. Oropharyngeal tumors were 43%, oral cavity 40% and laryngeal 17% of cases. Concurrent systemic therapy included platinum and cetuximab in 72% and 28%, respectively. Optimal TPT was found to be < 100 days with significantly better OS (P = 0.002) and RFS (P = 0.043) compared to TPT ≥100 days. On multivariate analysis; TPT<100 days, extracapsular nodal extension, high-risk score, lymphovascular space and perineural invasion were independent predictors for worse OS (P < 0.05). T4, extracapsular nodal extension and high-risk score were all significantly detrimental to RFS (P < 0.05). CONCLUSIONS: Addition of concomitant systemic therapy to adjuvant radiotherapy did not compensate for longer TPT in head and neck squamous cell carcinoma. Multidisciplinary coordinated care must be provided to ensure the early start of CRT with minimal treatment breaks.

Impact of CT reconstruction algorithm on auto-segmentation performance.

Model-based iterative reconstruction (MBIR) reduces CT imaging dose while maintaining image quality. However, MBIR reduces noise while preserving edges which may impact intensity-based tasks such as auto-segmentation. This work evaluates the sensitivity of an auto-contouring prostate atlas across multiple MBIR reconstruction protocols and benchmarks the results against filtered back projection (FBP). Images were created from raw projection data for 11 prostate cancer cases using FBP and nine different MBIR reconstructions (3 protocols/3 noise reduction levels) yielding 10 reconstructions/patient. Five bony structures, bladder, rectum, prostate, and seminal vesicles (SVs) were segmented using an auto-segmentation pipeline that renders 3D binary masks for analysis. Performance was evaluated for volume percent difference (VPD) and Dice similarity coefficient (DSC), using FBP as the gold standard. Nonparametric Friedman tests plus post hoc all pairwise comparisons were employed to test for significant differences (P < 0.05) for soft tissue organs and protocol/level combinations. A physician performed qualitative grading of 396 MBIR contours across the prostate, bladder, SVs, and rectum in comparison to FBP using a six-point scale. MBIR contours agreed with FBP for bony anatomy (DSC ≥ 0.98), bladder (DSC ≥ 0.94, VPD < 8.5%), and prostate (DSC = 0.94 ± 0.03, VPD = 4.50 ± 4.77% (range: 0.07-26.39%). Increased variability was observed for rectum (VPD = 7.50 ± 7.56% and DSC = 0.90 ± 0.08) and SVs (VPD and DSC of 8.23 ± 9.86% range (0.00-35.80%) and 0.87 ± 0.11, respectively). Over the all protocol/level comparisons, a significant difference was observed for the prostate VPD between BSPL1 and BSTL2 (adjusted P-value = 0.039). Nevertheless, 300 of 396 (75.8%) of the four soft tissue structures using MBIR were graded as equivalent or better than FBP, suggesting that MBIR offered potential improvements in auto-segmentation performance when compared to FBP. Future work may involve tuning organ-specific MBIR parameters to further improve auto-segmentation performance. Running title: Impact of CT Reconstruction Algorithm on Auto-segmentation Performance.

Impact of smoking on pathological features in oral cavity squamous cell carcinoma.

OBJECTIVES: We sought to determine whether smokers with oral cavity squamous cell carcinoma (OCSCC) have tumors with more adverse pathological features than in nonsmokers and whether or not these are predictive of outcomes. MATERIALS AND METHODS: We retrospectively identified 163 patients with American Joint Committee on Cancer stages I-IVa OCSCC diagnosed between 2005 and 2015 and treated with curative intent. A pathological risk score (PRS) was calculated using the National Comprehensive Cancer Network adverse risk factors: positive margin, extracapsular extension of lymph node metastases, pT3 or pT4 primary, N2 or N3 nodal disease, perineural invasion, and lymphovascular space invasion. Multivariable models were constructed to determine the independent predictors of overall survival (OS), recurrence-free survival (RFS), and PRS. RESULTS: A total of 108 (66.26%) were smokers and 55 nonsmokers. Three-year actuarial OS and RFS were 62% and 68% in smokers and 81% and 69% in nonsmokers, respectively (P = 0.06 and P = 0.63). Smokers were more likely to have advanced disease stage and tumors with aggressive pathological features than nonsmokers. Smokers had significantly worse PRS (mean ± standard deviation; 2.38 ± 2.19, median; 2.00) than nonsmokers (0.89 ± 1.21, 0.00) (P < 0.001). Older age, higher PRS, and smoking status were independent predictors of OS. Smoking or PRS did not predict for worse RFS. On multivariate analysis, independent predictors of PRS were smoking status and grade (P < 0.001). CONCLUSION: In patients with OCSCC, smokers have more aggressive disease as evidenced by more adverse pathological features than nonsmokers. Moreover, smoking is an independent predictor of OS but not RFS. The PRS is a significant predictor of OS and needs validation in the future studies.

Cardiac Substructure Segmentation and Dosimetry Using a Novel Hybrid Magnetic Resonance and Computed Tomography Cardiac Atlas.

PURPOSE: Radiation dose to the heart and cardiac substructures has been linked to cardiotoxicities. Because cardiac substructures are poorly visualized on treatment-planning computed tomography (CT) scans, we used the superior soft-tissue contrast of magnetic resonance (MR) imaging to optimize a hybrid MR/CT atlas for substructure dose assessment using CT. METHODS AND MATERIALS: Thirty-one patients with left-sided breast cancer underwent a T2-weighted MR imaging scan and noncontrast simulation CT scans. A radiation oncologist delineated 13 substructures (chambers, great vessels, coronary arteries, etc) using MR/CT information via cardiac-confined rigid registration. Ground-truth contours for 20 patients were inputted into an intensity-based deformable registration atlas and applied to 11 validation patients. Automatic segmentations involved using majority vote and Simultaneous Truth and Performance Level Estimation (STAPLE) strategies with 1 to 15 atlas matches. Performance was evaluated via Dice similarity coefficient (DSC), mean distance to agreement, and centroid displacement. Three physicians evaluated segmentation performance via consensus scoring by using a 5-point scale. Dosimetric assessment included measurements of mean heart dose, left ventricular volume receiving 5 Gy, and left anterior descending artery mean and maximum doses. RESULTS: Atlas approaches performed similarly well, with 7 of 13 substructures (heart, chambers, ascending aorta, and pulmonary artery) having DSC >0.75 when averaged over 11 validation patients. Coronary artery segmentations were not successful with the atlas-based approach (mean DSC <0.3). The STAPLE method with 10 matches yielded the highest DSC and the lowest mean distance to agreement for all high-performing substructures (omitting coronary arteries). For the STAPLE method with 10 matches, >50% of all validation contours had centroid displacements <3.0 mm, with the largest shifts in the coronary arteries. Atlas-generated contours had no statistical difference from ground truth for left anterior descending artery maximum dose, mean heart dose, and left ventricular volume receiving 5 Gy (P > .05). Qualitative contour grading showed that 8 substructures required minor modifications. CONCLUSIONS: The hybrid MR/CT atlas provided reliable segmentations of chambers, heart, and great vessels for patients undergoing noncontrast CT, suggesting potential widespread applicability for routine treatment planning.

Retroperitoneal Metastasis Abutting Small Bowel: A Novel Magnetic Resonance-Guided Radiation Approach.

Stereotactic body radiation therapy (SBRT) is an option for selected patients with metastatic disease. However, sometimes these lesions are located in such close proximity to critical normal structures that the use of safe tumoricidal SBRT doses is not achievable. Here we present a case in which real-time imaging and tracking with a magnetic resonance linear accelerator (MR-LINAC) provided a novel treatment approach and enabled safe treatment of the tumor using SBRT. Our case is a 69-year-old female who presented with localized recurrent small cell lung cancer with a retroperitoneal (FDG-avid) soft tissue lesion measuring 2.4 x 4.1 cm that was causing pain and right hydronephrosis. A Food and Drug Administration (FDA)-approved MR-LINAC system was utilized for planning and the delivery of 21 Gy in three fractions to the retroperitoneal lesion planning target volume (PTV), limited by the neighboring small bowel tolerance. The gross tumor volume (GTV) itself received 27 Gy (9 Gy per fraction). Simulation was performed using a volumetric MR imaging study in treatment position co-registered to a 4D-computed tomography (CT) image set for contouring of the target and organs at risk (OAR). Treatment planning was performed using the primary CT dataset. We developed a reasonable SBRT treatment plan to deliver the prescribed dose without exceeding tolerance doses to the right kidney, the small bowel and all other OAR's. Real-time MR imaging and tracking during treatment delivery enabled assessment of respiratory-induced target movement in relation to the small bowel and kidney. Gating was performed to halt treatment when PTV movement exceeded the 2-mm range as specified by the treating physician. The treatment course was concluded successfully. The patient denied any acute gastrointestinal or genitourinary toxicity. The pain was significantly improved within a short time following treatment. Follow-up CT showed a near complete response of the mass with total restoration of renal functions, allowing the ureteric stent to be removed. This response has been maintained for five months till the last follow-up. In conclusion, MR-guided planning and delivery using real-time MR imaging and tracking facilitated the treatment of the retroperitoneal mass accurately and efficiently with excellent clinical and radiological response and minimal to no toxicity. We would not discern it safe to treat this mass utilizing SBRT without this ability to accurately visualize the tumor boundary using magnetic resonance imaging (MRI), and offer tracking of the target within the millimeter of surrounding critical OAR's.