Increasing the efficiency of cone-beam CT based delta-radiomics using automated contours to predict radiotherapy-related toxicities in prostate cancer.

Extracting longitudinal image quantitative data, known as delta-radiomics, has the potential to capture changes in a patient's anatomy throughout the course of radiation treatment for prostate cancer. Some of the major challenges of delta-radiomics studies are contouring the structures for individual fractions and accruing patients' data in an efficient manner. The manual contouring process is often time consuming and would limit the efficiency of accruing larger sample sizes for future studies. The problem is amplified because the contours are often made by highly trained radiation oncologists with limited time to dedicate to research studies of this nature. This work compares the use of automated prostate contours generated using a deformable image-based algorithm to make predictive models of genitourinary and changes in total international prostate symptom score in comparison to manually contours for a cohort of fifty patients. Area under the curve of manual and automated models were compared using the Delong test. This study demonstrated that the delta-radiomics models were similar for both automated and manual delta-radiomics models.

Uncovering prostate cancer aggressiveness signal in T2-weighted MRI through a three-reference tissues normalization technique.

Quantitative T2-weighted MRI (T2W) interpretation is impeded by the variability of acquisition-related features, such as field strength, coil type, signal amplification, and pulse sequence parameters. The main purpose of this work is to develop an automated method for prostate T2W intensity normalization. The procedure includes the following: (i) a deep learning-based network utilizing MASK R-CNN for automatic segmentation of three reference tissues: gluteus maximus muscle, femur, and bladder; (ii) fitting a spline function between average intensities in these structures and reference values; and (iii) using the function to transform all T2W intensities. The T2W distributions in the prostate cancer regions of interest (ROIs) and normal appearing prostate tissue (NAT) were compared before and after normalization using Student's t-test. The ROIs' T2W associations with the Gleason Score (GS), Decipher genomic score, and a three-tier prostate cancer risk were evaluated with Spearman's correlation coefficient (rS ). T2W differences in indolent and aggressive prostate cancer lesions were also assessed. The MASK R-CNN was trained with manual contours from 32 patients. The normalization procedure was applied to an independent MRI dataset from 83 patients. T2W differences between ROIs and NAT significantly increased after normalization. T2W intensities in 231 biopsy ROIs were significantly negatively correlated with GS (rS = -0.21, p = 0.001), Decipher (rS = -0.193, p = 0.003), and three-tier risk (rS = -0.235, p < 0.001). The average T2W intensities in the aggressive ROIs were significantly lower than in the indolent ROIs after normalization. In conclusion, the automated triple-reference tissue normalization method significantly improved the discrimination between prostate cancer and normal prostate tissue. In addition, the normalized T2W intensities of cancer exhibited a significant association with tumor aggressiveness. By improving the quantitative utilization of the T2W in the assessment of prostate cancer on MRI, the new normalization method represents an important advance over clinical protocols that do not include sequences for the measurement of T2 relaxation times.

Clinical-Genomic Risk Group Classification of Suspicious Lesions on Prostate Multiparametric-MRI.

The utilization of multi-parametric MRI (mpMRI) in clinical decisions regarding prostate cancer patients' management has recently increased. After biopsy, clinicians can assess risk using National Comprehensive Cancer Network (NCCN) risk stratification schema and commercially available genomic classifiers, such as Decipher. We built radiomics-based models to predict lesions/patients at low risk prior to biopsy based on an established three-tier clinical-genomic classification system. Radiomic features were extracted from regions of positive biopsies and Normally Appearing Tissues (NAT) on T2-weighted and Diffusion-weighted Imaging. Using only clinical information available prior to biopsy, five models for predicting low-risk lesions/patients were evaluated, based on: 1: Clinical variables; 2: Lesion-based radiomic features; 3: Lesion and NAT radiomics; 4: Clinical and lesion-based radiomics; and 5: Clinical, lesion and NAT radiomic features. Eighty-three mpMRI exams from 78 men were analyzed. Models 1 and 2 performed similarly (Area under the receiver operating characteristic curve were 0.835 and 0.838, respectively), but radiomics significantly improved the lesion-based performance of the model in a subset analysis of patients with a negative Digital Rectal Exam (DRE). Adding normal tissue radiomics significantly improved the performance in all cases. Similar patterns were observed on patient-level models. To the best of our knowledge, this is the first study to demonstrate that machine learning radiomics-based models can predict patients' risk using combined clinical-genomic classification.

ED visits, hospital admissions and treatment breaks in head/neck cancer patients undergoing radiotherapy.

Objectives: Radiation therapy (RT) is an integral part of treatment of head/neck cancer (HNC) but is associated with many toxicities. We sought to evaluate sociodemographic, pathologic, and clinical factors associated with emergency department (ED) visits, hospital admissions (HA), and RT breaks in HNC patients undergoing curative-intent RT. Methods: We completed a Level 3 (Oxford criteria for evidence-based medicine) analysis of a cohort of HNC patients who underwent curative-intent RT at our institution from 2013 to 2017. We collected demographic characteristics and retrospectively assessed for heavy opioid use, ED visits or HA during RT as well as RT breaks. Treatment breaks were defined as total days to RT fractions ratio ≥1.6. Multivariable stepwise logistic regression analyses were done to determine the association of various sociodemographic, pathologic, and clinical characteristics with ED visits, HA and RT treatment breaks. Results: The cohort included 376 HNC patients (294 male, 82 female, median age 61). On multivariable analysis, significant factors associated with ED visits during RT were heavy opioid use and black race. Receipt of concomitant chemotherapy was the only factor associated with hospital admissions during RT. Advanced age, lower socioeconomic class, glandular site, and receipt of chemotherapy were all independently associated with RT breaks. Lower cancer stage and lack of substance abuse history were independently associated with lack of treatment breaks. Conclusion: HNC patients with factors such as heavy opioid use, Black race, receipt of concomitant chemotherapy, and lower socioeconomic class may require closer monitoring during RT.

Delineation of recurrent glioblastoma by whole brain spectroscopic magnetic resonance imaging.

BACKGROUND: Glioblastoma (GBM) cellularity correlates with whole brain spectroscopic MRI (sMRI) generated relative choline to N-Acetyl-Aspartate ratio (rChoNAA) mapping. In recurrent GBM (rGBM), tumor volume (TV) delineation is challenging and rChoNAA maps may assist with re-RT targeting. METHODS: Fourteen rGBM patients underwent sMRI in a prospective study. Whole brain sMRI was performed to generate rChoNAA maps. TVs were delineated by the union of rChoNAA ratio over 2 (rChoNAA > 2) on sMRI and T1PC. rChoNAA > 2 volumes were compared with multiparametric MRI sequences including T1PC, T2/FLAIR, diffusion-restriction on apparent diffusion coefficient (ADC) maps, and perfusion relative cerebral blood volume (rCBV). RESULTS: rChoNAA > 2 (mean 27.6 cc, range 6.6-79.1 cc) was different from other imaging modalities (P ≤ 0.05). Mean T1PC volumes were 10.7 cc (range 1.2-31.4 cc). The mean non-overlapping volume of rChoNAA > 2 and T1PC was 29.2 cm3. rChoNAA > 2 was 287% larger (range 23% smaller-873% larger) than T1PC. T2/FLAIR volumes (mean 111.7 cc, range 19.0-232.7 cc) were much larger than other modalities. rCBV volumes (mean 6.2 cc, range 0.2-19.1 cc) and ADC volumes were tiny (mean 0.8 cc, range 0-3.7 cc). Eight in-field failures were observed. Three patients failed outside T1PC but within rChoNAA > 2. No grade 3 toxicities attributable to re-RT were observed. Median progression-free and overall survival for re-RT patients were 6.5 and 7.1 months, respectively. CONCLUSIONS: Treatment of rGBM may be optimized by sMRI, and failure patterns suggest benefit for dose-escalation within sMRI-delineated volumes. Dose-escalation and radiologic-pathologic studies are underway to confirm the utility of sMRI in rGBM.

Cone-beam CT delta-radiomics to predict genitourinary toxicities and international prostate symptom of prostate cancer patients: a pilot study.

For prostate cancer (PCa) patients treated with definitive radiotherapy (RT), acute and late RT-related genitourinary (GU) toxicities adversely impact disease-specific quality of life. Early warning of potential RT toxicities can prompt interventions that may prevent or mitigate future adverse events. During intensity modulated RT (IMRT) of PCa, daily cone-beam computed tomography (CBCT) images are used to improve treatment accuracy through image guidance. This work investigated the performance of CBCT-based delta-radiomic features (DRF) models to predict acute and sub-acute International Prostate Symptom Scores (IPSS) and Common Terminology Criteria for Adverse Events (CTCAE) version 5 GU toxicity grades for 50 PCa patients treated with definitive RT. Delta-radiomics models were built using logistic regression, random forest for feature selection, and a 1000 iteration bootstrapping leave one analysis for cross validation. To our knowledge, no prior studies of PCa have used DRF models based on daily CBCT images. AUC of 0.83 for IPSS and greater than 0.7 for CTCAE grades were achieved as early as week 1 of treatment. DRF extracted from CBCT images showed promise for the development of models predictive of RT outcomes. Future studies will include using artificial intelligence and machine learning to expand CBCT sample sizes available for radiomics analysis.

Longitudinal Changes and Predictive Value of Multiparametric MRI Features for Prostate Cancer Patients Treated with MRI-Guided Lattice Extreme Ablative Dose (LEAD) Boost Radiotherapy.

We investigated the longitudinal changes in multiparametric MRI (mpMRI) (T2-weighted, Apparent Diffusion Coefficient (ADC), and Dynamic Contrast Enhanced (DCE-)MRI) of prostate cancer patients receiving Lattice Extreme Ablative Dose (LEAD) radiotherapy (RT) and the capability of their imaging features to predict RT outcome based on endpoint biopsies. Ninety-five mpMRI exams from 25 patients, acquired pre-RT and at 3-, 9-, and 24-months post-RT were analyzed. MRI/Ultrasound-fused biopsies were acquired pre- and at two-years post-RT (endpoint). Five regions of interest (ROIs) were analyzed: Gross tumor volume (GTV), normally-appearing tissue (NAT) and peritumoral volume in both peripheral (PZ) and transition (TZ) zones. Diffusion and perfusion radiomics features were extracted from mpMRI and compared before and after RT using two-tailed Student t-tests. Selected features at the four scan points and their differences (Δ radiomics) were used in multivariate logistic regression models to predict the endpoint biopsy positivity. Baseline ADC values were significantly different between GTV, NAT-PZ, and NAT-TZ (p-values < 0.005). Pharmaco-kinetic features changed significantly in the GTV at 3-month post-RT compared to baseline. Several radiomics features at baseline and three-months post-RT were significantly associated with endpoint biopsy positivity and were used to build models with high predictive power of this endpoint (AUC = 0.98 and 0.89, respectively). Our study characterized the RT-induced changes in perfusion and diffusion. Quantitative imaging features from mpMRI show promise as being predictive of endpoint biopsy positivity.

Survival and Yield of Surveillance Imaging in Long-Term Survivors of Brain Metastasis Treated with Stereotactic Radiosurgery.

OBJECTIVES: The optimal frequency of surveillance brain magnetic resonance imaging (MRI) in long-term survivors with brain metastases after stereotactic radiosurgery (SRS) is unknown. Our aim was to identify the optimal frequency of surveillance imaging in long-term survivors with brain metastases after SRS. METHODS: Eligible patients were identified from a cohort treated with SRS definitively or postoperatively at our institution from 2014 to 2019 with no central nervous system (CNS) failure within 12 months from SRS. Time to CNS disease failure diagnosis and cost per patient were estimated using theoretical MRI schedules of 2, 3, 4, and 6 months starting 1 year after SRS until CNS failure. Time to diagnosis was calculated from the date of CNS progression to the theoretical imaging date on each schedule. RESULTS: This cohort included 55 patients (median follow-up from SRS: 2.48 years). During the study period, 20.0% had CNS disease failure (median: 2.26 years from SRS treatment). In this cohort, a theoretical 2-month, 3-month, 4-month, and 6-month MRI brain surveillance schedule produced a respective estimated time to diagnosis of CNS disease failure of 1.11, 1.74, 1.65, and 3.65 months. The cost of expedited diagnosis for the cohort (dollars/month) for each theoretical imaging schedule compared with a 6-month surveillance schedule was $6600 for a 2-month protocol, $4496 for a 3-month protocol, and $2180 for a 4-month protocol. CONCLUSIONS: Based on cost-benefit, a 4-month MRI brain schedule should be considered in patients with metastatic disease to the brain treated definitively or postoperatively with SRS without evidence of CNS recurrence at 1 year.

Clinicogenomic characterization of prostate cancer liver metastases.

BACKGROUND: The site of prostate cancer metastasis is an important predictor of oncologic outcomes, however, the clinicogenomic characteristics associated with the site are not well-defined. Herein, we characterize the genomic alterations associated with the metastatic site of prostate cancer. METHODS: We analyzed clinical and genomic data from prostate cancer patients with metastatic disease and known metastatic sites from publicly available targeted sequencing data. RESULTS: Prostate cancer metastasis to the liver versus other sites of metastasis conferred a high hazard for death in patients with metastatic prostate cancer (HR: 3.96, 95% CI: 2.4-6.5, p < 0.0001). Genomic analysis of metastatic tissues of prostate cancer-specific genes demonstrated that liver metastases were more enriched with MYC amplification (29.5% vs. 9.8%, FDR = 0.001), PTEN deletion (42% vs. 20.8%, FDR = 0.005), and PIK3CB amplification (8.2% vs. 0.9, FDR = 0.005) compared to other sites. No point mutations were significantly associated with liver metastasis compared to other metastatic sites. CONCLUSION: Liver metastases in prostate cancer are associated with poor survival and aggressive genomic features, including MYC-amplification, PTEN-deletion, and PIK3CB-amplification. These findings could have prognostic, treatment, and trial implications.

Integrating Prostate-specific Antigen Kinetics into Contemporary Predictive Nomograms of Salvage Radiotherapy After Radical Prostatectomy.

BACKGROUND: Salvage radiotherapy (SRT) is an established treatment for men with biochemical recurrence following radical prostatectomy (RP). There are several risk factors associated with adverse outcomes; however, the value of postoperative prostate-specific antigen (PSA) kinetics is less clear in the ultrasensitive PSA era. OBJECTIVE: To characterize the impact of PSA kinetics on outcomes following SRT and generate nomograms to aid in identifying patients with an increased risk of adverse clinical outcomes. DESIGN, SETTING, AND PARTICIPANTS: A multi-institutional analysis was conducted of 1005 patients with prostate cancer treated with SRT after RP, with a median follow-up of 5 years. OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: Variables examined include immediate postoperative PSA, postoperative PSA doubling time (DT), and pre-SRT PSA, in addition to previously identified predictive factors. Multivariable survival analyses were completed using Fine-Gray competing risk regression. Rates of biochemical failure (BF), distant metastasis (DM), and prostate cancer-specific mortality (PCSM) were estimated by the cumulative incidence method. Nomograms were generated from multivariable competing risk regression with bootstrap cross-validation. RESULTS AND LIMITATIONS: Factors associated with BF after SRT include PSA DT <6 mo, initial postoperative PSA ≥0.2 ng/ml, higher pre-SRT PSA, lack of androgen deprivation therapy, a higher Gleason score (GS), negative margins, seminal vesicle invasion, lack of pelvic nodal radiation, radiation total dose <66 Gy, a longer RP to SRT interval, and older age (p < 0.05 for each). Factors associated with DM include PSA DT <6 mo, pre-SRT PSA, a higher GS, and negative margins. Factors associated with PCSM include PSA DT not calculable or <6 mo and a higher GS. Nomograms were generated to estimate the risks of BF (concordance index [CI] 0.74), DM (CI 0.77), and PCSM (CI 0.77). Limitations include retrospective nature, broad treatment eras, institutional variations, and multiple methods available for the estimation of PSA DT. CONCLUSIONS: Postoperative PSA kinetics, particularly pre-SRT PSA and PSA DT, are strongly associated with adverse oncologic outcomes following SRT and should be considered in management decisions. PATIENT SUMMARY: In this report of men with prostate cancer who developed a prostate-specific antigen (PSA) recurrence after prostatectomy, we found that PSA levels after surgery and how quickly a PSA level doubles significantly impact the chance of prostate cancer recurrence after salvage radiation therapy. Based on this information, we created a tool to calculate a man's chance of cancer recurrence after salvage radiation therapy, and these estimations can be used to discuss whether additional treatment with radiation should be considered.

Assessment of CT to CBCT contour mapping for radiomic feature analysis in prostate cancer.

This study provides a quantitative assessment of the accuracy of a commercially available deformable image registration (DIR) algorithm to automatically generate prostate contours and additionally investigates the robustness of radiomic features to differing contours. Twenty-eight prostate cancer patients enrolled on an institutional review board (IRB) approved protocol were selected. Planning CTs (pCTs) were deformably registered to daily cone-beam CTs (CBCTs) to generate prostate contours (auto contours). The prostate contours were also manually drawn by a physician. Quantitative assessment of deformed versus manually drawn prostate contours on daily CBCT images was performed using Dice similarity coefficient (DSC), mean distance-to-agreement (MDA), difference in center-of-mass position (ΔCM) and difference in volume (ΔVol). Radiomic features from 6 classes were extracted from each contour. Lin's concordance correlation coefficient (CCC) and mean absolute percent difference in radiomic feature-derived data (mean |%Δ|RF) between auto and manual contours were calculated. The mean (± SD) DSC, MDA, ΔCM and ΔVol between the auto and manual prostate contours were 0.90 ± 0.04, 1.81 ± 0.47 mm, 2.17 ± 1.26 mm and 5.1 ± 4.1% respectively. Of the 1,010 fractions under consideration, 94.8% of DIRs were within TG-132 recommended tolerance. 30 radiomic features had a CCC > 0.90 and 21 had a mean |%∆|RF < 5%. Auto-propagation of prostate contours resulted in nearly 95% of DIRs within tolerance recommendations of TG-132, leading to the majority of features being regarded as acceptably robust. The use of auto contours for radiomic feature analysis is promising but must be done with caution.

Novel genomic signature predictive of response to immune checkpoint blockade: A pan-cancer analysis from project Genomics Evidence Neo-plasia Information Exchange (GENIE).

BACKGROUND: High tumor mutation burden (TMB) and total mutation count (TMC) can be predictive of better response to immune checkpoint blockade (ICB). Nevertheless, TMB and TMC are limited by variation across cancers and inconsistent definitions due to different profiling methods (targeted vs whole genome sequencing). Our objective was to identify genomic alterations (GAs) associated with ICB response and builds a novel genomic signature predictive of ICB response, independent of TMB/TMC. METHODS: This was a pan-cancer next generation sequencing (NGS)-association study using January 2014-May 2016 data from AACR Project Genomics Evidence Neo-plasia Information Exchange (GENIE). Participants included 6619 patients with metastatic or un-resectable cancer across 9 cancer types (including 1572 ICB-treated patients). GA data was collected using next-generation sequencing (NGS) assays and downloaded from cbioportal.org. Predictive analyses for ICB response were performed to develop the signature (ImmGA). RESULTS: GAs in 16 genes were associated with improved OS in ICB-treated patients (p < 0.005). 13 GAs were associated with an OS benefit in ICB-treated patients (Pinteraction < 0.05); these genes composed the ImmGA signature. High ImmGA score (≥2 alterations out of 13 predictive GAs) was associated with better OS in ICB-treated patients (AHR:0.67, 95%CI [0.6-0.75], p = 1.4e-12), even after accounting for TMC (Pinteraction = 8e-16). High ImmGA was associated with better OS in ICB-treated patients across most cancers and across different ICB treatment modalities. CONCLUSION: A novel signature predictive of ICB response (ImmGA) was developed from 13 GAs. Further investigation of the utility of ImmGA for treatment and trial selection is warranted.

Opioid use patterns in patients with head and neck cancer receiving radiation therapy: Single-institution retrospective analysis characterizing patients who did not require opioid therapy.

BACKGROUND: We had previously analyzed the variables that determine the rates of opioid use at 1-year postradiotherapy in patients with head and neck cancer. Here we analyze the variables associated with opioid abstinence during and in the 12 months after radiotherapy at our institution. METHODS: We identified a cohort of patients with head and neck cancer who received radiotherapy as part of curative treatment at our institution. Logistic regression analyses were performed to determine socioeconomic and clinical factors associated with opioid abstinence. RESULTS: The cohort included 376 patients. On multivariable analysis, patients from an upper-income class (p = 0.004), black race (p = 0.004), older (p = 0.008), with dependent children (p < 0.001) or receiving surgery (p = 0.002) were more likely to abstain from opioids, while patients using analgesic mouthwash (p = 0.009) or higher pain scale (p = 0.002) were less likely. CONCLUSION: Socioeconomic and treatment characteristics are associated with opioid abstinence during and following radiation treatment in patients with head and neck cancer.

Repeatability of CBCT radiomic features and their correlation with CT radiomic features for prostate cancer.

PURPOSE: Radiomic features of cone-beam CT (CBCT) images have potential as biomarkers to predict treatment response and prognosis for patients of prostate cancer. Previous studies of radiomic feature analysis for prostate cancer were assessed in a variety of imaging modalities, including MRI, PET, and CT, but usually limited to a pretreatment setting. However, CBCT images may provide an opportunity to capture early morphological changes to the tumor during treatment that could lead to timely treatment adaptation. This work investigated the quality of CBCT-based radiomic features and their relationship with reconstruction methods applied to the CBCT projections and the preprocessing methods used in feature extraction. Moreover, CBCT features were correlated with planning CT (pCT) features to further assess the viability of CBCT radiomic features. METHODS: The quality of 42 CBCT-based radiomic features was assessed according to their repeatability and reproducibility. Repeatability was quantified by correlating radiomic features between 20 CBCT scans that also had repeated scans within 15 minutes. Reproducibility was quantified by correlating radiomic features between the planning CT (pCT) and the first fraction CBCT for 20 patients. Concordance correlation coefficients (CCC) of radiomic features were used to estimate the repeatability and reproducibility of radiomic features. The same patient dataset was assessed using different reconstruction methods applied to the CBCT projections. CBCT images were generated using 18 reconstruction methods using iterative (iCBCT) and standard (sCBCT) reconstructions, three convolution filters, and five noise suppression filters. Eighteen preprocessing settings were also considered. RESULTS: Overall, CBCT radiomic features were more repeatable than reproducible. Five radiomic features are repeatable in > 97% of the reconstruction and preprocessing methods, and come from the gray-level size zone matrix (GLSZM), neighborhood gray-tone difference matrix (NGTDM), and gray-level-run length matrix (GLRLM) radiomic feature classes. These radiomic features were reproducible in > 9.8% of the reconstruction and preprocessing methods. Noise suppression and convolution filter smoothing increased radiomic features repeatability, but decreased reproducibility. The top-repeatable iCBCT method (iCBCT-Sharp-VeryHigh) is more repeatable than the top-repeatable sCBCT method (sCBCT-Smooth) in 64% of the radiomic features. CONCLUSION: Methods for reconstruction and preprocessing that improve CBCT radiomic feature repeatability often decrease reproducibility. The best approach may be to use methods that strike a balance repeatability and reproducibility such as iCBCT-Sharp-VeryLow-1-Lloyd-256 that has 17 repeatable and eight reproducible radiomic features. Previous radiomic studies that only used pCT radiomic features have generated prognostic models of prostate cancer outcome. Since our study indicates that CBCT radiomic features correlated well with a subset of pCT radiomic features, one may expect CBCT radiomics to also generate prognostic models for prostate cancer.

The role of radiomics in prostate cancer radiotherapy.

"Radiomics," as it refers to the extraction and analysis of a large number of advanced quantitative radiological features from medical images using high-throughput methods, is perfectly suited as an engine for effectively sifting through the multiple series of prostate images from before, during, and after radiotherapy (RT). Multiparametric (mp)MRI, planning CT, and cone beam CT (CBCT) routinely acquired throughout RT and the radiomics pipeline are developed for extraction of thousands of variables. Radiomics data are in a format that is appropriate for building descriptive and predictive models relating image features to diagnostic, prognostic, or predictive information. Prediction of Gleason score, the histopathologic cancer grade, has been the mainstay of the radiomic efforts in prostate cancer. While Gleason score (GS) is still the best predictor of treatment outcome, there are other novel applications of quantitative imaging that are tailored to RT. In this review, we summarize the radiomics efforts and discuss several promising concepts such as delta-radiomics and radiogenomics for utilizing image features for assessment of the aggressiveness of prostate cancer and its outcome. We also discuss opportunities for quantitative imaging with the advance of instrumentation in MRI-guided therapies.

Segmentation of prostate and prostate zones using deep learning : A multi-MRI vendor analysis.

PURPOSE: Develop a deep-learning-based segmentation algorithm for prostate and its peripheral zone (PZ) that is reliable across multiple MRI vendors. METHODS: This is a retrospective study. The dataset consisted of 550 MRIs (Siemens-330, General Electric[GE]-220). A multistream 3D convolutional neural network is used for automatic segmentation of the prostate and its PZ using T2-weighted (T2-w) MRI. Prostate and PZ were manually contoured on axial T2­w. The network uses axial, coronal, and sagittal T2­w series as input. The preprocessing of the input data includes bias correction, resampling, and image normalization. A dataset from two MRI vendors (Siemens and GE) is used to test the proposed network. Six different models were trained, three for the prostate and three for the PZ. Of the three, two were trained on data from each vendor separately, and a third (Combined) on the aggregate of the datasets. The Dice coefficient (DSC) is used to compare the manual and predicted segmentation. RESULTS: For prostate segmentation, the Combined model obtained DSCs of 0.893 ± 0.036 and 0.825 ± 0.112 (mean ± standard deviation) on Siemens and GE, respectively. For PZ, the best DSCs were from the Combined model: 0.811 ± 0.079 and 0.788 ± 0.093. While the Siemens model underperformed on the GE dataset and vice versa, the Combined model achieved robust performance on both datasets. CONCLUSION: The proposed network has a performance comparable to the interexpert variability for segmenting the prostate and its PZ. Combining images from different MRI vendors on the training of the network is of paramount importance for building a universal model for prostate and PZ segmentation.

Phase I Trial of MRI-Guided Prostate Cancer Lattice Extreme Ablative Dose (LEAD) Boost Radiation Therapy.

PURPOSE: A phase I clinical trial was designed to test the feasibility and toxicity of administering high-dose spatially fractionated radiation therapy to magnetic resonance imaging (MRI)-defined prostate tumor volumes, in addition to standard treatment. METHODS AND MATERIALS: We enrolled 25 men with favorable to high-risk prostate cancer and 1 to 3 suspicious multiparametric MRI (mpMRI) gross tumor volumes (GTVs). The mpMRI-GTVs were treated on day 1 with 12 to 14 Gy via dose cylinders using a lattice extreme ablative dose technique. The entire prostate, along with the proximal seminal vesicles, was then treated to 76 Gy at 2 Gy/fraction. For some high-risk patients, the distal seminal vesicles and pelvic lymph nodes received 56 Gy at 1.47 Gy/fraction concurrently in 38 fractions. The total dose to the lattice extreme ablative dose cylinder volume(s) was 88 to 90 Gy (112-123 Gy in 2.0 Gy equivalents, assuming an α-to-ß ratio of 3). RESULTS: Dosimetric parameters were satisfactorily met. Median follow-up was 66 months. There were no grade 3 acute/subacute genitourinary or gastrointestinal adverse events. Maximum late genitourinary toxicity was grade 1 in 15 (60%), grade 2 in 4 (16%), and grade 4 in 1 (4%; sepsis after a posttreatment transurethral resection). Maximum late gastrointestinal toxicity was grade 1 in 11 (44%) and grade 2 in 4 (16%). Two patients experienced biochemical failure. CONCLUSIONS: External beam radiation therapy delivered with an upfront spatially fractionated, stereotactic high-dose mpMRI-GTV boost is feasible and was not associated with any unexpected events. The technique is now part of a follow-up phase II randomized trial.

Long-term opioid use in curative-intent radiotherapy: One-Year outcomes in head/neck cancer patients.

BACKGROUND: No study has determined the incidence of long-term opioid use, or risk factors for long-term use, ≥1 year after radiotherapy. METHODS: Medical records of 276 head/neck cancer patients were retrospectively assessed for persistent opioid use 1-year after curative-intent radiotherapy. Numerous potential risk factors were assessed and the physicians' documented reasons for continued use were qualitatively categorized as suspected opioid use disorder (OUD) or as medically indicated for control of ongoing pain. RESULTS: Of note, 20 of 276 patients continued using opioids long-term. High maximum opioid dose and the use of opioids and/or psychotropics/non-opioid analgesics at the radiation oncology intake visit were associated with this outcome. Three patients continued due to suspected OUD and 17 due to medical indications. CONCLUSION: Of note, 7.2% of patients developed long-term opioid use, which was associated with high maximum opioid dose and early initiation of opioids and/or psychotropics/non-opioid analgesics. Physicians cited medical indications as the primary reason for continued use.

Dosimetric comparison of intensity-modulated radiation therapy for early-stage glottic cancers with and without the air cavity in the planning target volume.

For early-stage glottic cancers, intensity-modulated radiation therapy (IMRT) has been shown to have comparable local control to 3D-conformal radiotherapy with the advantage of decreased dose to the carotid arteries. The planning target volume (PTV) for early glottic cancers typically includes the entire larynx, plus a 3 to 5 mm uniform margin. The air cavity within the larynx creates a challenge for the inverse optimization process as the software attempts to "build up" dose within the air. This unnecessary attempt at dose build-up in air can lead to hot spots within the rest of the PTV and surrounding soft tissue. We hypothesized that removal of the air from the PTV would decrease hot spots and allow for a more homogeneous plan while still maintaining adequate coverage of the PTV. We analyzed 20 consecutive patients with early-stage glottic cancer, T1-2N0, who received IMRT at our institution from April 2015 to December 2016. Each patient received 63 to 65.25 Gy in 2.25 Gy per fraction. Two plans were created for each case: one in which the PTV included the laryngeal air cavity and one in which the air cavity was subtracted from the PTV to create a new PTV-air structure. Dosimetric variables were collected for PTV-air structure from both IMRT plans, including V100%, D98% D2%, and D0.2%. Dosimetric variables for spinal cord and the carotid arteries were also recorded. Homogeneity index (HI) defined as D98/D2 was calculated. Two-sided t-tests were used to compare dosimetric variables. The median PTV volume was 69.9 cc (standard deviation [SD] ± 28.7 cc) and the median air cavity volume removed was 11.0 cc (SD ± 3.4 cc). A 2-sided t-test revealed a statistically significant decrease in max dose (112.7% vs 108.8%, p value = 0.0002) and improvement of HI (0.93 vs 0.91, p value = 0.0023) for the PTV air in the IMRT plan optimized for PTV air, which had air excluded, compared to the IMRT plan optimized for PTV with air included. There was no significant worsening of PTV-air coverage or significant increase in doses to the organs at risk (OARs). The removal of the air cavity from the PTV for early-stage glottic cancers does not compromise PTV coverage or sparing of OARs and can result in a more homogeneous IMRT plan. A more homogeneous plan has the potential to reduce treatment morbidity, although further study is warranted to investigate the clinical impact of air cavity removal from the PTV.