Investigation of autosegmentation techniques on T2-weighted MRI for off-line dose reconstruction in MR-linac workflow for head and neck cancers.

BACKGROUND: In order to accurately accumulate delivered dose for head and neck cancer patients treated with the Adapt to Position workflow on the 1.5T magnetic resonance imaging (MRI)-linear accelerator (MR-linac), the low-resolution T2-weighted MRIs used for daily setup must be segmented to enable reconstruction of the delivered dose at each fraction. PURPOSE: In this pilot study, we evaluate various autosegmentation methods for head and neck organs at risk (OARs) on on-board setup MRIs from the MR-linac for off-line reconstruction of delivered dose. METHODS: Seven OARs (parotid glands, submandibular glands, mandible, spinal cord, and brainstem) were contoured on 43 images by seven observers each. Ground truth contours were generated using a simultaneous truth and performance level estimation (STAPLE) algorithm. Twenty total autosegmentation methods were evaluated in ADMIRE: 1-9) atlas-based autosegmentation using a population atlas library (PAL) of 5/10/15 patients with STAPLE, patch fusion (PF), random forest (RF) for label fusion; 10-19) autosegmentation using images from a patient's 1-4 prior fractions (individualized patient prior [IPP]) using STAPLE/PF/RF; 20) deep learning (DL) (3D ResUNet trained on 43 ground truth structure sets plus 45 contoured by one observer). Execution time was measured for each method. Autosegmented structures were compared to ground truth structures using the Dice similarity coefficient, mean surface distance (MSD), Hausdorff distance (HD), and Jaccard index (JI). For each metric and OAR, performance was compared to the inter-observer variability using Dunn's test with control. Methods were compared pairwise using the Steel-Dwass test for each metric pooled across all OARs. Further dosimetric analysis was performed on three high-performing autosegmentation methods (DL, IPP with RF and 4 fractions [IPP_RF_4], IPP with 1 fraction [IPP_1]), and one low-performing (PAL with STAPLE and 5 atlases [PAL_ST_5]). For five patients, delivered doses from clinical plans were recalculated on setup images with ground truth and autosegmented structure sets. Differences in maximum and mean dose to each structure between the ground truth and autosegmented structures were calculated and correlated with geometric metrics. RESULTS: DL and IPP methods performed best overall, all significantly outperforming inter-observer variability and with no significant difference between methods in pairwise comparison. PAL methods performed worst overall; most were not significantly different from the inter-observer variability or from each other. DL was the fastest method (33 s per case) and PAL methods the slowest (3.7-13.8 min per case). Execution time increased with a number of prior fractions/atlases for IPP and PAL. For DL, IPP_1, and IPP_RF_4, the majority (95%) of dose differences were within ± 250 cGy from ground truth, but outlier differences up to 785 cGy occurred. Dose differences were much higher for PAL_ST_5, with outlier differences up to 1920 cGy. Dose differences showed weak but significant correlations with all geometric metrics (R2 between 0.030 and 0.314). CONCLUSIONS: The autosegmentation methods offering the best combination of performance and execution time are DL and IPP_1. Dose reconstruction on on-board T2-weighted MRIs is feasible with autosegmented structures with minimal dosimetric variation from ground truth, but contours should be visually inspected prior to dose reconstruction in an end-to-end dose accumulation workflow.

Insurance Denial of Care for Randomized Controlled Trial-Eligible Patients: Incidence and Success Rate of Peer-To-Peer Authorization in Allowing Patients to Remain Trial-Eligible.

INTRODUCTION: Insurance denials for clinical trials serve as a pertinent barrier for patients to remain trial-eligible, thus hindering the development of therapies and the overall advancement of health care. We present results from an ongoing oncology randomized clinical trial regarding insurance denials and peer-to-peer authorization (P2PA) success rate in allowing patients to remain trial-eligible. METHODS: The ongoing Spine Patient Optimal Radiosurgery Treatment for Symptomatic Metastatic Neoplasms Phase II trial randomizes spine cancer patients to treatment with spine radiosurgery/stereotactic body radiation therapy (SBRT) versus conventional external beam radiation therapy (EBRT). Trial-eligible patients during the first 3 months of enrollment are examined to determine whether the option of SBRT was denied by their insurance. Advocacy for overcoming SBRT denial in P2PA centered on SBRT being recommended as a preferred treatment modality in the National Comprehensive Cancer Network guidelines, and the recent level I evidence demonstrating the advantages of SBRT over EBRT for symptomatic spine cancer. RESULTS: Of 15 trial-eligible patients, 3 (20%) experienced insurance denials for SBRT. P2PA resulted in the reversal of denials in all 3 patients, allowing each to remain trial-eligible for randomization between SBRT and cEBRT. CONCLUSIONS: Despite a clinical oncologic treatment modality for which recent Level 1 evidence is available, the insurance denial rate was 20%. A vigilant P2PA strategy focusing on highlighting National Comprehensive Cancer Network guidelines and the supporting Level 1 evidence resulted in a very high rate of reversing initial denial.

Contemporary trends in management of uveal melanoma.

Purpose: In the management of uveal melanoma, eye plaque brachytherapy (EPBT) has replaced enucleation as the standard of care for small size tumors that require treatment, and for medium size tumors. In the modern era, EPBT is being utilized more frequently for certain large tumors as well. While there is prospective randomized evidence to support utilization of EPBT for tumors of appropriate dimensions, it is unclear what the actual practice patterns are across the United States. The purpose of this publication was to look at contemporary trends in the management of uveal melanoma across the United States to determine whether practices are appropriately adopting EPBT, and to investigate demographic and socio-economic factors that might be associated with deviations from this standard of care. Material and methods: The National Cancer Database was queried (2004-2015) for patients with uveal melanoma. Data regarding tumor characteristics and treatment were collected. Two-sided Pearson χ2 test was used to compare categorical frequencies between patients who received globe preserving treatments vs. those who received enucleation. Multivariable logistic regression modeling was used to determine characteristics predictive for receiving enucleation. Results: The enucleation rate for small/medium tumors (≤ 10 mm apical height and ≤ 16 mm basal diameter) decreased from 20% in 2004 to 10% in 2015. The EPBT rate for large tumors increased from 30% in 2004 to 45% in 2015. Numerous demographic and socio-economic factors were found to be associated with higher rates of enucleation. Conclusions: The overall trend across the nation is a decreased enucleation rate for small/medium tumors, and an increased EPBT rate for large tumors. A fraction of patients who should be candidates for EPBT are instead receiving enucleation, and in this study, we have shown that certain adverse demographic factors are associated with this.

Eye plaque brachytherapy versus enucleation for ocular melanoma: an analysis from the National Cancer Database.

PURPOSE: There is no current randomized data comparing the efficacy of brachytherapy and enucleation for patients with larger sized tumors. The purpose of the present study was to use a large, contemporary database to determine current practice patterns and compare survival outcomes between different management options for patients with choroidal melanoma of various sizes. MATERIAL AND METHODS: The National Cancer Database was queried (2004-2014) for histologically-confirmed choroidal melanoma for patients treated with brachytherapy versus enucleation. Chi-square test was used to compare categorized demographic and clinical variables in both arms. Kaplan-Meier analysis evaluated overall survival (OS). Cox proportional hazards assessment determined variables associated with OS. Patients were divided into cohorts representing small, medium, and large tumors. Propensity scores matching (PSM) was utilized to compare more similar cohorts. RESULTS: A total of 7,096 patients met the selection criteria; 5,501 (78%) patients received brachytherapy and 1,595 (22%) patients were treated with enucleation. After PSM, 5-yr OS for small tumors was 87% vs. 64%, for medium tumors was 77% vs. 57%, and for large tumors was 68% vs. 46% for brachytherapy and enucleation, respectively (p < 0.001). Following PSM, multivariate Cox regression found older age (hazard ratio [HR] = 1.76, 95% confidence interval [CI] = 1.51-2.06), more comorbidities (HR = 1.46, 95% CI = 1.25-1.70), extraocular extension (EOE) (HR = 1.25, 95% CI = 1.06-1.48), ciliary body invasion (CBI) (HR = 1.20, 95% CI = 1.02-1.40), and larger size (HR = 1.52, 95% CI = 1.40-1.66) were negative prognosticators of survival. Brachytherapy was a positive prognosticator of survival (HR = 0.45, 95% CI = 0.40-0.51). CONCLUSIONS: Patients selected for brachytherapy had improved survival compared to enucleation in all size cohorts. EOE and CBI are significantly higher in the enucleation cohort and are important negative prognosticators for survival selected against patients having brachytherapy. Brachytherapy is a reasonable treatment option for certain patients with large size tumors.