Pilot study of radiation oncology peer review in low middle income country (LMIC) through cloud-based platform.

PURPOSE: Peer review is an essential step in clinical quality assurance for radiation therapy. There are very little data on peer reviews from low-middle-income countries (LMIC). With increasing access to advanced technologies in LMIC also, peer review is becoming more important to ensure quality and standard of care. We evaluated cloud-based e-Peer review in our network of cancer centers in India with an aim to study its feasibility and impact on care. MATERIALS AND METHODS: Four out of 13 cancer centers across India were selected for this pilot study. All team members were trained on the e-Peer review platform before the initiation of the study. A lead dosimetrist from a centralized planning site was selected to share new cases every week. Cases treated with only definitive intent were selected. The link to the cases was sent through an email to reviewing physicians. The following aspects were reviewed for each case. 1) Work up and staging. 2) Treatment intent and prescription. 3) Target contours. 4) Normal organ at risk contours. 5) Dose-volume-histogram (DVH) with clinical goals attached. Cases were marked as "Not Appropriate," "Appropriate," "Appropriate with minor finding," and "Represent with major revisions" as per volume and plan review. RESULTS: Over a period of 3 months, 100 cases underwent peer review before the start of treatment. Median turnover time was 48 (interquartile range: 24-96) hours. The median time for review was 8 min with time to review cases requiring major and minor changes being 12 and 6 min, respectively (P < 0.001). Of all the cases reviewed, no changes, minor changes, and major changes were suggested for 36%, 48%, and 16% of cases, respectively. The most frequent reason for major changes was contouring corrections (15%). Also, 31.3% of major changes underwent recontouring and replanning before initiation of treatment. CONCLUSION: Peer review was feasible in our setting through this cloud-based peer review system, with median turnover time and time taken for review being 48 h and 8 min, respectively. Like published data from the Western world, peer review led to changes that could impact patient care delivery and outcome. We plan to implement this across the remaining centers in our network.

Future of Telemedicine in Radiation Oncology.

Telemedicine allows providers and patients to communicate without being in the same room through video platforms or telephone. Like the increased use of telework for businesses, telemedicine exploded during the pandemic. While many workplaces and clinics have returned to some level of in-person interactions, the convenience and comfort have given telemedicine staying power. Patients can be seen from the comfort of their homes; family members can join from the same or a different location. Driving, obtaining childcare, or taking time off from work is unnecessary. Pediatric patients' parents can pull them into the conversation at appropriate times and avoid the awkwardness of having them leave for portions of the discussion. Because virtual visits are more efficient for everyone, they can often be scheduled sooner than an in-person visit. While not every visit can be done without the patient physically with the provider, many can. This is particularly true for cancer patients, who often have several visits with multiple providers. For immunocompromised patients, there is an added benefit of avoiding exposure from travel and a hospital visit. Oncology and radiation oncology practices have widely adopted telemedicine. While legal and logistical barriers exist in some areas of the world, these are sure to be resolved to make this medicine feasible for all in the modern era.

Frailty assessment in geriatric radiation oncology.

Older adults with cancer are at risk of over-treatment or under-treatment, and treatment decision-making is difficult due to both the complexity of adverse aging and under-representation in clinical trials. It is recommended to perform a frailty assessment before treatment decision-making. Although the importance of radiotherapy increases in geriatric oncology, there is less evidence base information on frailty assessment in radiation oncology than in medical/surgical oncology. The present literature review analyzed the available data regarding frailty assessment tools in geriatric radiation oncology. The predictive value of geriatric assessment on survival outcomes has been shown in many cancer subtypes treated with radiotherapy. Additionally, the Geriatric-8 score is the most evidenced screening tool in frailty assessment. However, researches are ongoing on the cut-off points of geriatric screening tools and which one is the best. Prospective randomized controlled trials are required for the integration of geriatric screening tools and geriatric assessment-driven interventions into geriatric radiation oncology practice.

The Role of Artificial Intelligence on Tumor Boards: Perspectives from Surgeons, Medical Oncologists and Radiation Oncologists.

The integration of multidisciplinary tumor boards (MTBs) is fundamental in delivering state-of-the-art cancer treatment, facilitating collaborative diagnosis and management by a diverse team of specialists. Despite the clear benefits in personalized patient care and improved outcomes, the increasing burden on MTBs due to rising cancer incidence and financial constraints necessitates innovative solutions. The advent of artificial intelligence (AI) in the medical field offers a promising avenue to support clinical decision-making. This review explores the perspectives of clinicians dedicated to the care of cancer patients-surgeons, medical oncologists, and radiation oncologists-on the application of AI within MTBs. Additionally, it examines the role of AI across various clinical specialties involved in cancer diagnosis and treatment. By analyzing both the potential and the challenges, this study underscores how AI can enhance multidisciplinary discussions and optimize treatment plans. The findings highlight the transformative role that AI may play in refining oncology care and sustaining the efficacy of MTBs amidst growing clinical demands.

Large Language Models to Help Appeal Denied Radiotherapy Services.

PURPOSE: Large language model (LLM) artificial intelligences may help physicians appeal insurer denials of prescribed medical services, a task that delays patient care and contributes to burnout. We evaluated LLM performance at this task for denials of radiotherapy services. METHODS: We evaluated generative pretrained transformer 3.5 (GPT-3.5; OpenAI, San Francisco, CA), GPT-4, GPT-4 with internet search functionality (GPT-4web), and GPT-3.5ft. The latter was developed by fine-tuning GPT-3.5 via an OpenAI application programming interface with 53 examples of appeal letters written by radiation oncologists. Twenty test prompts with simulated patient histories were programmatically presented to the LLMs, and output appeal letters were scored by three blinded radiation oncologists for language representation, clinical detail inclusion, clinical reasoning validity, literature citations, and overall readiness for insurer submission. RESULTS: Interobserver agreement between radiation oncologists' scores was moderate or better for all domains (Cohen's kappa coefficients: 0.41-0.91). GPT-3.5, GPT-4, and GPT-4web wrote letters that were on average linguistically clear, summarized provided clinical histories without confabulation, reasoned appropriately, and were scored useful to expedite the insurance appeal process. GPT-4 and GPT-4web letters demonstrated superior clinical reasoning and were readier for submission than GPT-3.5 letters (P < .001). Fine-tuning increased GPT-3.5ft confabulation and compromised performance compared with other LLMs across all domains (P < .001). All LLMs, including GPT-4web, were poor at supporting clinical assertions with existing, relevant, and appropriately cited primary literature. CONCLUSION: When prompted appropriately, three commercially available LLMs drafted letters that physicians deemed would expedite appealing insurer denials of radiotherapy services. LLMs may decrease this task's clerical workload on providers. However, LLM performance worsened when fine-tuned with a task-specific, small training data set.

Advancing the Collaboration Between Imaging and Radiation Oncology.

The fusion of cutting-edge imaging technologies with radiation therapy (RT) has catalyzed transformative breakthroughs in cancer treatment in recent decades. It is critical for us to review our achievements and preview into the next phase for future synergy between imaging and RT. This paper serves as a review and preview for fostering collaboration between these two domains in the forthcoming decade. Firstly, it delineates ten prospective directions ranging from technological innovations to leveraging imaging data in RT planning, execution, and preclinical research. Secondly, it presents major directions for infrastructure and team development in facilitating interdisciplinary synergy and clinical translation. We envision a future where seamless integration of imaging technologies into RT will not only meet the demands of RT but also unlock novel functionalities, enhancing accuracy, efficiency, safety, and ultimately, the standard of care for patients worldwide.

A method for empirically validating FMEA RPN scores in a radiation oncology clinic using physics QC data.

In failure modes and effects analysis (FMEA), the components of the risk priority number (RPN) for a failure mode (FM) are often chosen by consensus. We describe an empirical method for estimating the occurrence (O) and detectability (D) components of a RPN. The method requires for a given FM that its associated quality control measure be performed twice as is the case when a FM is checked for in an initial physics check and again during a weekly physics check. If instances of the FM caught by these checks are recorded, O and D can be computed. Incorporation of the remaining RPN component, Severity, is discussed. This method can be used as part of quality management design ahead of an anticipated FMEA or afterwards to validate consensus values.

Which Modality of SFRT Should be Considered First for Bulky Tumor Radiation Therapy, GRID or LATTICE?

Spatially fractionated radiation therapy (SFRT), also known as the GRID and LATTICE radiotherapy (GRT, LRT), the concept of treating tumors by delivering a spatially modulated dose with highly non-uniform dose distributions, is a treatment modality of growing interest in radiation oncology, physics, and radiation biology. Clinical experience in SFRT has suggested that GRID and LATTICE therapy can achieve a high response and low toxicity in the treatment of refractory and bulky tumors. Limited initially to GRID therapy using block collimators, advanced, and versatile multi-leaf collimators, volumetric modulated arc technologies and particle therapy have since increased the capabilities and individualization of SFRT and expanded the clinical investigation of SFRT to various dosing regimens, multiple malignancies, tumor types and sites. As a 3D modulation approach outgrown from traditional 2D GRID, LATTICE therapy aims to reconfigure the traditional SFRT as spatial modulation of the radiation is confined solely to the tumor volume. The distinctively different beam geometries used in LATTICE therapy have led to appreciable variations in dose-volume distributions, compared to GRID therapy. The clinical relevance of the variations in dose-volume distribution between LATTICE and traditional GRID therapies is a crucial factor in determining their adoption in clinical practice. In this Point-Counterpoint contribution, the authors debate the pros and cons of GRID and LATTICE therapy. Both modalities have been used in clinics and their applicability and optimal use have been discussed in this article.

Autologous Peripheral Blood Stem Cell Transplantation and Timing of Local Radiation Therapy in Patients With Malignant Lymphoma: Results of the Japanese Radiation Oncology Study Group Questionnaire Survey.

BACKGROUND/AIM: Malignant lymphoma (ML) including Hodgkin's lymphoma and non-Hodgkin's lymphoma is often treated with local radiation therapy (RT) in combination with autologous hematopoietic stem cell transplantation (ASCT) to prevent relapse; however, the efficacy and optimal timing of this approach is unclear. In this study, a national survey conducted by the Japanese Radiation Oncology Study Group reviewed ML cases from 2011 to 2019 to determine whether RT should be added to ASCT, focusing on the use of autologous peripheral blood stem cell transplantation (auto-PBSCT), a predominant form of ASCT. PATIENTS AND METHODS: The survey encompassed 92 patients from 11 institutes, and assessed histological ML types, treatment regimens, timing of RT relative to auto-PBSCT, and associated adverse events. RESULTS: The results indicated no significant differences in adverse events, including myelosuppression, based on the timing of RT in relation to auto-PBSCT. However, anemia was more prevalent when RT was administered before auto-PBSCT, and there was a higher incidence of neutropenia recovery delay in patients receiving RT after auto-PBSCT. CONCLUSION: This study provides valuable insights into the variable practices of auto-PBSCT and local RT in ML treatment, emphasizing the need for optimized timing of these therapies to improve patient outcomes and reduce complications.

Artificial intelligence solution to accelerate the acquisition of MRI images: Impact on the therapeutic care in oncology in radiology and radiotherapy departments.

PURPOSE: MRI is essential in the management of brain tumours. However, long waiting times reduce patient accessibility. Reducing acquisition time could improve access but at the cost of spatial resolution and diagnostic quality. A commercially available artificial intelligence (AI) solution, SubtleMR™, can increase the resolution of acquired images. The objective of this prospective study was to evaluate the impact of this algorithm that halves the acquisition time on the detectability of brain lesions in radiology and radiotherapy. MATERIAL AND METHODS: The T1/T2 MRI of 33 patients with brain metastases or meningiomas were analysed. Images acquired quickly have a matrix divided by two which halves the acquisition time. The visual quality and lesion detectability of the AI images were evaluated by radiologists and radiation oncologist as well as pixel intensity and lesions size. RESULTS: The subjective quality of the image is lower for the AI images compared to the reference images. However, the analysis of lesion detectability shows a specificity of 1 and a sensitivity of 0.92 and 0.77 for radiology and radiotherapy respectively. Undetected lesions on the IA image are lesions with a diameter less than 4mm and statistically low average gadolinium-enhancement contrast. CONCLUSION: It is possible to reduce MRI acquisition times by half using the commercial algorithm to restore the characteristics of the image and obtain good specificity and sensitivity for lesions with a diameter greater than 4mm.

Associations Between Radiation Oncologist Demographic Factors and Segmentation Similarity Benchmarks: Insights From a Crowd-Sourced Challenge Using Bayesian Estimation.

PURPOSE: The quality of radiotherapy auto-segmentation training data, primarily derived from clinician observers, is of utmost importance. However, the factors influencing the quality of clinician-derived segmentations are poorly understood; our study aims to quantify these factors. METHODS: Organ at risk (OAR) and tumor-related segmentations provided by radiation oncologists from the Contouring Collaborative for Consensus in Radiation Oncology data set were used. Segmentations were derived from five disease sites: breast, sarcoma, head and neck (H&N), gynecologic (GYN), and GI. Segmentation quality was determined on a structure-by-structure basis by comparing the observer segmentations with an expert-derived consensus, which served as a reference standard benchmark. The Dice similarity coefficient (DSC) was primarily used as a metric for the comparisons. DSC was stratified into binary groups on the basis of structure-specific expert-derived interobserver variability (IOV) cutoffs. Generalized linear mixed-effects models using Bayesian estimation were used to investigate the association between demographic variables and the binarized DSC for each disease site. Variables with a highest density interval excluding zero were considered to substantially affect the outcome measure. RESULTS: Five hundred seventy-four, 110, 452, 112, and 48 segmentations were used for the breast, sarcoma, H&N, GYN, and GI cases, respectively. The median percentage of segmentations that crossed the expert DSC IOV cutoff when stratified by structure type was 55% and 31% for OARs and tumors, respectively. Regression analysis revealed that the structure being tumor-related had a substantial negative impact on binarized DSC for the breast, sarcoma, H&N, and GI cases. There were no recurring relationships between segmentation quality and demographic variables across the cases, with most variables demonstrating large standard deviations. CONCLUSION: Our study highlights substantial uncertainty surrounding conventionally presumed factors influencing segmentation quality relative to benchmarks.

Navigating the Critical Translational Questions for Implementing FLASH in the Clinic.

The "FLASH effect" is an increased therapeutic index, that is, reduced normal tissue toxicity for a given degree of anti-cancer efficacy, produced by ultra-rapid irradiation delivered on time scales orders of magnitude shorter than currently conventional in the clinic for the same doses. This phenomenon has been observed in numerous preclinical in vivo tumor and normal tissue models. While the underlying biological mechanism(s) remain to be elucidated, a path to clinical implementation of FLASH can be paved by addressing several critical translational questions. Technological questions pertinent to each beam type (eg, electron, proton, photon) also dictate the logical progression of experimentation required to move forward in safe and decisive clinical trials. Here we review the available preclinical data pertaining to these questions and how they may inform strategies for FLASH cancer therapy clinical trials.

Deformable registration of magnetic resonance images using unsupervised deep learning in neuro-/radiation oncology.

PURPOSE: Accurate deformable registration of magnetic resonance imaging (MRI) scans containing pathologies is challenging due to changes in tissue appearance. In this paper, we developed a novel automated three-dimensional (3D) convolutional U-Net based deformable image registration (ConvUNet-DIR) method using unsupervised learning to establish correspondence between baseline pre-operative and follow-up MRI scans of patients with brain glioma. METHODS: This study involved multi-parametric brain MRI scans (T1, T1-contrast enhanced, T2, FLAIR) acquired at pre-operative and follow-up time for 160 patients diagnosed with glioma, representing the BraTS-Reg 2022 challenge dataset. ConvUNet-DIR, a deep learning-based deformable registration workflow using 3D U-Net style architecture as a core, was developed to establish correspondence between the MRI scans. The workflow consists of three components: (1) the U-Net learns features from pairs of MRI scans and estimates a mapping between them, (2) the grid generator computes the sampling grid based on the derived transformation parameters, and (3) the spatial transformation layer generates a warped image by applying the sampling operation using interpolation. A similarity measure was used as a loss function for the network with a regularization parameter limiting the deformation. The model was trained via unsupervised learning using pairs of MRI scans on a training data set (n = 102) and validated on a validation data set (n = 26) to assess its generalizability. Its performance was evaluated on a test set (n = 32) by computing the Dice score and structural similarity index (SSIM) quantitative metrics. The model's performance also was compared with the baseline state-of-the-art VoxelMorph (VM1 and VM2) learning-based algorithms. RESULTS: The ConvUNet-DIR model showed promising competency in performing accurate 3D deformable registration. It achieved a mean Dice score of 0.975 ± 0.003 and SSIM of 0.908 ± 0.011 on the test set (n = 32). Experimental results also demonstrated that ConvUNet-DIR outperformed the VoxelMorph algorithms concerning Dice (VM1: 0.969 ± 0.006 and VM2: 0.957 ± 0.008) and SSIM (VM1: 0.893 ± 0.012 and VM2: 0.857 ± 0.017) metrics. The time required to perform a registration for a pair of MRI scans is about 1 s on the CPU. CONCLUSIONS: The developed deep learning-based model can perform an end-to-end deformable registration of a pair of 3D MRI scans for glioma patients without human intervention. The model could provide accurate, efficient, and robust deformable registration without needing pre-alignment and labeling. It outperformed the state-of-the-art VoxelMorph learning-based deformable registration algorithms and other supervised/unsupervised deep learning-based methods reported in the literature.

Radiation Oncology Ransomware Attack Response Risk Analysis Using Failure Modes and Effects Analysis.

PURPOSE: There have been numerous significant ransomware attacks impacting Radiation Oncology in the past 5 years. Research into ransomware attack response in Radiation Oncology has consisted of case reports and descriptive articles and has lacked quantitative studies. The purpose of this work was to identify the significant safety risks to patients being treated with radiation therapy during a ransomware attack scenario, using Failure Modes and Effects Analysis. METHODS AND MATERIALS: A multi-institutional and multidisciplinary team conducted a Failure Modes and Effects Analysis by developing process maps and using Risk Priority Number (RPN) scores to quantify the increased likelihood of incidents in a ransomware attack scenario. The situation that was simulated was a ransomware attack that had removed the capability to access the Record and Verify (R&V) system. Five situations were considered: 1) a standard treatment of a patient with and without an R&V, 2) a standard treatment of a patient for the first fraction right after the R&V capabilities are disabled, and 3) 3 situations in which a plan modification was required. RPN scores were compared with and without R&V functionality. RESULTS: The data indicate that RPN scores increased by 71% (range, 38%-96%) when R&V functionality is disabled compared with a nonransomware attack state where R&V functionality is available. The failure modes with the highest RPN in the simulated ransomware attack state included incorrectly identifying patients on treatment, incorrectly identifying where a patient is in their course of treatment, treating the incorrect patient, and incorrectly tracking delivered fractions. CONCLUSIONS: The presented study quantifies the increased risk of incidents when treating in a ransomware attack state, identifies key failure modes that should be prioritized when preparing for a ransomware attack, and provides data that can be used to guide future ransomware resiliency research.

Radiomics for clinical decision support in radiation oncology.

Radiomics is a promising tool for the development of quantitative biomarkers to support clinical decision-making. It has been shown to improve the prediction of response to treatment and outcome in different settings, particularly in the field of radiation oncology by optimising the dose delivery solutions and reducing the rate of radiation-induced side effects, leading to a fully personalised approach. Despite the promising results offered by radiomics at each of these stages, standardised methodologies, reproducibility and interpretability of results are still lacking, limiting the potential clinical impact of these tools. In this review, we briefly describe the principles of radiomics and the most relevant applications of radiomics at each stage of cancer management in the framework of radiation oncology. Furthermore, the integration of radiomics into clinical decision support systems is analysed, defining the challenges and offering possible solutions for translating radiomics into a clinically applicable tool.

Optimizing Outpatient Radiation Oncology Consult Workflow by Using Time-Driven Activity-Based Costing: Efficiency and Financial Impacts.

PURPOSE: Clinical efficiency is a key component of value-based health care. Our objective here was to identify workflow inefficiencies by using time-driven activity-based costing (TDABC) and evaluate the implementation of a new clinical workflow in high-volume outpatient radiation oncology clinics. METHODS: Our quality improvement study was conducted with the Departments of GI, Genitourinary (GU), and Thoracic Radiation Oncology at a large academic cancer center and four community network sites. TDABC was used to create process maps and optimize workflow for outpatient consults. Patient encounter metrics were captured with a real-time status function in the electronic medical record. Time metrics were compared using Mann-Whitney U tests. RESULTS: Individual patient encounter data for 1,328 consults before the intervention and 1,234 afterward across all sections were included. The median overall cycle time was reduced by 21% in GI (19 minutes), 18% in GU (16 minutes), and 12% at the community sites (9 minutes). The median financial savings per consult were $52 in US dollars (USD) for the GI, $33 USD for GU, $30 USD for thoracic, and $42 USD for the community sites. Patient satisfaction surveys (from 127 of 228 patients) showed that 99% of patients reported that their providers spent adequate time with them and 91% reported being seen by a care provider in a timely manner. CONCLUSION: TDABC can effectively identify opportunities to improve clinical efficiency. Implementing workflow changes on the basis of our findings led to substantial reductions in overall encounter cycle times across several departments, as well as high patient satisfaction and significant financial savings.