A national approach for automated collection of standardized and population-based radiation therapy data in Sweden

Benchmarking a Foundation Large Language Model on its Ability to Relabel Structure Names in Accordance With the American Association of Physicists in Medicine Task Group-263 Report

PURPOSE

To introduce the concept of using large language models (LLMs) to relabel structure names in accordance with the American Association of Physicists in Medicine Task Group-263 standard and to establish a benchmark for future studies to reference.

METHODS AND MATERIALS

Generative Pretrained Transformer (GPT)-4 was implemented within a Digital Imaging and Communications in Medicine server. Upon receiving a structure-set Digital Imaging and Communications in Medicine file, the server prompts GPT-4 to relabel the structure names according to the American Association of Physicists in Medicine Task Group-263 report. The results were evaluated for 3 disease sites: prostate, head and neck, and thorax. For each disease site, 150 patients were randomly selected for manually tuning the instructions prompt (in batches of 50), and 50 patients were randomly selected for evaluation. Structure names considered were those that were most likely to be relevant for studies using structure contours for many patients.

RESULTS

The per-patient accuracy was 97.2%, 98.3%, and 97.1% for prostate, head and neck, and thorax disease sites, respectively. On a per-structure basis, the clinical target volume was relabeled correctly in 100%, 95.3%, and 92.9% of cases, respectively.

CONCLUSIONS

Given the accuracy of GPT-4 in relabeling structure names as presented in this work, LLMs are poised to become an important method for standardizing structure names in radiation oncology, especially considering the rapid advancements in LLM capabilities that are likely to continue.

Towards a European prospective data registry for particle therapy

The evidence for the value of particle therapy (PT) is still sparse. While randomized trials remain a cornerstone for robust comparisons with photon-based radiotherapy, data registries collecting real-world data can play a crucial role in building evidence for new developments. This Perspective describes how the European Particle Therapy Network (EPTN) is actively working on establishing a prospective data registry encompassing all patients undergoing PT in European centers. Several obstacles and hurdles are discussed, for instance harmonization of nomenclature and structure of technical and dosimetric data and data protection issues. A preferred approach is the adoption of a federated data registry model with transparent and agile governance to meet European requirements for data protection, transfer, and processing. Funding of the registry, especially for operation after the initial setup process, remains a major challenge.

Deep learning-based classification of organs at risk and delineation guideline in pelvic cancer radiation therapy

Deep learning (DL) models for radiation therapy (RT) image segmentation require accurately annotated training data. Multiple organ delineation guidelines exist; however, information on the used guideline is not provided with the delineation. Extraction of training data with coherent guidelines can therefore be challenging. We present a supervised classification method for pelvis structure delineations where bowel cavity, femoral heads, bladder, and rectum data, with two guidelines, were classified. The impact on DL-based segmentation quality using mixed guideline training data was also demonstrated. Bowel cavity was manually delineated on CT images for anal cancer patients (n = 170) according to guidelines Devisetty and RTOG. The DL segmentation quality from using training data with coherent or mixed guidelines was investigated. A supervised 3D squeeze-and-excite SENet-154 model was trained to classify two bowel cavity delineation guidelines. In addition, a pelvis CT dataset with manual delineations from prostate cancer patients (n = 1854) was used where data with an alternative guideline for femoral heads, rectum, and bladder were generated using commercial software. The model was evaluated on internal (n = 200) and external test data (n = 99). By using mixed, compared to coherent, delineation guideline training data mean DICE score decreased 3% units, mean Hausdorff distance (95%) increased 5 mm and mean surface distance (MSD) increased 1 mm. The classification of bowel cavity test data achieved 99.8% unweighted classification accuracy, 99.9% macro average precision, 97.2% macro average recall, and 98.5% macro average F1. Corresponding metrics for the pelvis internal test data were all 99% or above and for the external pelvis test data they were 96.3%, 96.6%, 93.3%, and 94.6%. Impaired segmentation performance was observed for training data with mixed guidelines. The DL delineation classification models achieved excellent results on internal and external test data. This can facilitate automated guideline-specific data extraction while avoiding the need for consistent and correct structure labels.

Driving quality improvement through better data: The story of New Zealand's radiation oncology collection

Aotearoa/New Zealand is one of the first nations in the world to develop a comprehensive, high-quality collection of radiation therapy data (the Radiation Oncology Collection, ROC) that is able to report on treatment delivery by health region, patient demographics and service provider. This has been guided by radiation therapy leaders, who have been instrumental in overseeing the establishment of clear and robust data definitions, a centralised database and outputs delivered via an online tool. In this paper, we detail the development of the ROC, provide examples of variation in practice identified from the ROC and how these changed over time, then consider the ramifications of the ROC in the wider context of cancer care quality improvement. In addition to a review of relevant literature, primary data were sourced from the ROC on radiation therapy provided nationally in New Zealand between 2017 and 2020. The total intervention rate, number of fractions and doses are reported for select cancers by way of examples of national variation in practice. Results from the ROC have highlighted areas of treatment variation and have prompted increased uptake of hypofractionation for curative prostate and breast cancer treatment and for palliation of bone metastases. Future development of the ROC will increase its use for quality improvement and ultimately link to a real time cancer services database.

Radiotherapy-specific quality indicators at national level: How to make it happen

PURPOSE /OBJECTIVE

To promote best practice and quality of care, the Belgian College of Physicians for Radiotherapy Centers established a set of radiotherapy specific quality indicators for benchmarking on a national level. This paper describes the development, the collected QIs, the observed trends and the departments' evaluation of this initiative.

MATERIAL AND METHODS

The Donabedian approach was used, focussing on structural, process and outcome QIs. The criteria for QI selection were availability, required for low-threshold regular collection, and applicability to guidelines and good practice. The QIs were collected yearly and individualized reports were sent out to all RT departments. In 2021, a national survey was held to evaluate the ease of data collection and submission, and the perceived importance and validity of the collected QIs.

RESULTS

18 structural QI and 37 process and outcome parameters (n = 25 patients/pathology/department) were collected. The participation rate amounted to 95 % overall. The analysis gave a national overview of RT activity, resources, clinical practice and reported acute toxicities. The individualized reports allowed departments to benchmark their performance. The 2021 survey indicated that the QIs were overall easy to collect, relevant and reliable. The collection of acute recorded toxicities was deemed a weak point due to inter-observer variabilities and lack of follow-up time.

CONCLUSION

QI collection on a national level is a valuable process in steering quality improvement initiatives. The feasibility and relevance was demonstrated with a high level of participation. The national initiative will continue to evolve as a quality monitoring and improvement tool.

Transparency in quality of radiotherapy for breast cancer in the Netherlands: a national registration of radiotherapy-parameters

Background

Radiotherapy (RT) is part of the curative treatment of approximately 70% of breast cancer (BC) patients. Wide practice variation has been reported in RT dose, fractionation and its treatment planning for BC. To decrease this practice variation, it is essential to first gain insight into the current variation in RT treatment between institutes. This paper describes the development of the NABON Breast Cancer Audit-Radiotherapy (NBCA-R), a structural nationwide registry of BC RT data of all BC patients treated with at least surgery and RT.

Methods

A working group consisting of representatives of the BC Platform of the Dutch Radiotherapy Society selected a set of dose volume parameters deemed to be surrogate outcome parameters, both for tumour control and toxicity. Two pilot studies were carried out in six RT institutes. In the first pilot study, data were manually entered into a secured web-based system. In the second pilot study, an automatic Digital Imaging and Communications in Medicine (DICOM) RT upload module was created and tested.

Results

The NBCA-R dataset was created by selecting RT parameters describing given dose, target volumes, coverage and homogeneity, and dose to organs at risk (OAR). Entering the data was made mandatory for all Dutch RT departments. In the first pilot study (N = 1093), quite some variation was already detected. Application of partial breast irradiation varied from 0 to 17% between the 6 institutes and boost to the tumour bed from 26.5 to 70.2%. For patients treated to the left breast or chest wall only, the average mean heart dose (MHD) varied from 0.80 to 1.82 Gy; for patients treated to the breast/chest wall only, the average mean lung dose (MLD) varied from 2.06 to 3.3 Gy. In the second pilot study 6 departments implemented the DICOM-RT upload module in daily practice. Anonymised data will be available for researchers via a FAIR (Findable, Accessible, Interoperable, Reusable) framework.

Conclusions

We have developed a set of RT parameters and implemented registration for all Dutch BC patients. With the use of an automated upload module registration burden will be minimized. Based on the data in the NBCA-R analyses of the practice variation will be done, with the ultimate aim to improve quality of BC RT.

Trial registration Retrospectively registered.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13014-022-02043-0.

Deep learning-based classification and structure name standardization for organ at risk and target delineations in prostate cancer radiotherapy

Radiotherapy (RT) datasets can suffer from variations in annotation of organ at risk (OAR) and target structures. Annotation standards exist, but their description for prostate targets is limited. This restricts the use of such data for supervised machine learning purposes as it requires properly annotated data. The aim of this work was to develop a modality independent deep learning (DL) model for automatic classification and annotation of prostate RT DICOM structures. Delineated prostate organs at risk (OAR), support- and target structures (gross tumor volume [GTV]/clinical target volume [CTV]/planning target volume [PTV]), along with or without separate vesicles and/or lymph nodes, were extracted as binary masks from 1854 patients. An image modality independent 2D InceptionResNetV2 classification network was trained with varying amounts of training data using four image input channels. Channel 1-3 consisted of orthogonal 2D projections from each individual binary structure. The fourth channel contained a summation of the other available binary structure masks. Structure classification performance was assessed in independent CT (n = 200 pat) and magnetic resonance imaging (MRI) (n = 40 pat) test datasets and an external CT (n = 99 pat) dataset from another clinic. A weighted classification accuracy of 99.4% was achieved during training. The unweighted classification accuracy and the weighted average F1 score among different structures in the CT test dataset were 98.8% and 98.4% and 98.6% and 98.5% for the MRI test dataset, respectively. The external CT dataset yielded the corresponding results 98.4% and 98.7% when analyzed for trained structures only, and results from the full dataset yielded 79.6% and 75.2%. Most misclassifications in the external CT dataset occurred due to multiple CTVs and PTVs being fused together, which was not included in the training data. Our proposed DL-based method for automated renaming and standardization of prostate radiotherapy annotations shows great potential. Clinic specific contouring standards however need to be represented in the training data for successful use. Source code is available at https://github.com/jamtheim/DicomRTStructRenamerPublic.

The power of empirical data; lessons from the clinical registry initiatives in Scandinavian cancer care

BACKGROUND

In Scandinavia, there is a strong tradition for research and quality monitoring based on registry data. In Denmark, Norway and Sweden, 63 clinical registries collect data on disease characteristics, treatment and outcome of various cancer diagnoses and groups based on process-related and outcome-related variables.

AIM

We describe the cancer-related clinical registries, compare organizational structures and quality indicators and provide examples of how these registries have been used to monitor clinical performance, develop prediction models, assess outcome and provide quality benchmarks. Further, we define unmet needs such as inclusion of patient-reported outcome variables, harmonization of variables and barriers for data sharing.

RESULTS AND CONCLUSIONS

The clinical registry framework provides an empirical basis for evidence-based development of high-quality and equitable cancer care. The registries can be used to follow implementation of new treatment principles and monitor patterns of care across geographical areas and patient groups. At the same time, the lessons learnt suggest that further developments and coordination are needed to utilize the full potential of the registry initiative in cancer care.

Radical radiotherapy for prostate cancer: patterns of care in Sweden 1998-2016

Introduction: Radiotherapy is an established treatment option for prostate cancer (PCa), both as primary treatment and secondary treatment after radical prostatectomy (RP). Since 1998, detailed data on radiotherapy delivered to Swedish men with PCa (e.g. treatment modalities, absorbed doses, fractionation) have been collated within PCa data Base Sweden (PCBaSe). This study reports patterns of radical radiotherapy for PCa in Sweden over the past two decades.Materials and methods: All men with non-metastatic PCa (1998-2016) who received external beam radiotherapy (EBRT) or high or low dose-rate brachytherapy (HDR-BT/LDR-BT) were identified in PCBaSe. Analyses included: trends in radiation techniques, fractionation patterns and total doses over time; PCa-specific survival comparing treatment in 2007-2017 with 1998-2006; and regional variation in type of primary radiotherapy.Results: About 20,876 men underwent primary radiotherapy. The main treatment modalities include conventionally fractionated (2.0 Gy/fraction) EBRT (51%), EBRT with HDR-BT boost (27%) and hypofractionated (>2.4 Gy/fraction) EBRT (11%). EBRT with photon or proton boost and HDR-BT and LDR-BT monotherapies were each used minimally. Use of dose-escalated EBRT (>74 Gy) and moderate hypofractionation increased over time, while use of HDR-BT declined. Considerable regional variation in treatment modalities was apparent. Risk of PCa death following primary radiotherapy had declined for intermediate-risk (HR: 0.60; 95%CI 0.47-0.87) and high-risk PCa (HR: 0.72; 95%CI 0.61-0.86).Discussion: Increased use of dose escalation and hypofractionated EBRT has occurred in Sweden over the past two decades, reflecting current evidence and practice guidelines. Disease-specific outcomes have also improved. Data collected in PCBaSe provide an excellent resource for further research into RT use in PCa management.

Initial experience with introducing national guidelines for CT- and MRI-based delineation of organs at risk in radiotherapy

A fundamental problem in radiotherapy is the variation of organ at risk (OAR) volumes. Here we present our initial experience in engaging a large Radiation Oncology (RO) community to agree on national guidelines for OAR delineations. Our project builds on associated standardization initiatives and invites professionals from all radiotherapy departments nationwide. Presently, one guideline (rectum) has successfully been agreed on by a majority vote. Reaching out to all relevant parties in a timely manner and motivating funding agencies to support the work represented early challenges. Population-based data and a scalable methodological approach are major strengths of the proposed strategy.

Retrospective analysis of breast radiotherapy treatment plans: Curating the 'non-curated'

INTRODUCTION

This paper provides a demonstration of how non-curated data can be retrospectively cleaned, so that existing repositories of radiotherapy treatment planning data can be used to complete bulk retrospective analyses of dosimetric trends and other plan characteristics.

METHODS

A non curated archive of 1137 radiotherapy treatment plans accumulated over a 12-month period, from five radiotherapy centres operated by one institution, was used to investigate and demonstrate a process of clinical data cleansing, by: identifying and translating inconsistent structure names; correcting inconsistent lung contouring; excluding plans for treatments other than breast tangents and plans without identifiable PTV, lung and heart structures; and identifying but not excluding plans that deviated from the local planning protocol. PTV, heart and lung dose-volume metrics were evaluated, in addition to a sample of personnel and linac load indicators.

RESULTS

Data cleansing reduced the number of treatment plans in the sample by 35.7%. Inconsistent structure names were successfully identified and translated (e.g. 35 different names for lung). Automatically separating whole lung structures into left and right lung structures allowed the effect of contralateral and ipsilateral lung dose to be evaluated, while introducing some small uncertainties, compared to manual contouring. PTV doses were indicative of prescription doses. Breast treatment work was unevenly distributed between oncologists and between metropolitan and regional centres.

CONCLUSION

This paper exemplifies the data cleansing and data analysis steps that may be completed using existing treatment planning data, to provide individual radiation oncology departments with access to information on their own patient populations. Clearly, the well-planned and systematic recording of new, high quality data is the preferred solution, but the retrospective curation of non-curated data may be a useful interim solution, for radiation oncology departments where the systems for recording of new data have yet to be designed and agreed.

A Systematic Post-QUANTEC Review of Tolerance Doses for Late Toxicity After Prostate Cancer Radiation Therapy

PURPOSE

The aims of this study were to systematically review tolerance doses for late distinct gastrointestinal (GI), genitourinary (GU), and sexual dysfunction (SD) symptoms after external beam radiation therapy (EBRT) alone and treatments involving brachytherapy (BT) for prostate cancer after Quantitative Analysis of Normal Tissue Effects in the Clinic (QUANTEC) and ultimately to perform quantitative syntheses of identified dose/volume tolerances represented by dose-volume histogram (DVH) thresholds, that is, statistically significant (P ≤ .05) cutoff points between symptomatic and asymptomatic patients in a certain study.

METHODS AND MATERIALS

PubMed was scrutinized for full-text articles in English after QUANTEC (January 1, 2010). The inclusion criteria were randomized controlled trials, case-control studies, or cohort studies with tolerance doses for late distinct symptoms ≥3 months after primary radiation therapy for prostate cancer (N > 30). All DVH thresholds were converted into equivalent doses in 2-Gy fractions (EQD2α/β) and were fitted with a linear or linear-quadratic function (goodness of fit, R2). The review was registered on PROSPERO (CRD42016042464).

RESULTS

From 33 identified studies, which included 36 to 746 patients per symptom domain, the majority of dose/volume tolerances were derived for GI toxicity after EBRT alone (GI, 97 thresholds; GU, 8 thresholds; SD, 1 threshold). For 5 symptoms (defecation urgency, diarrhea, fecal incontinence, proctitis, and rectal bleeding), relationships between dose/volume tolerances across studies (R2 = 0.93 [0.82-1.00]), and across symptoms, leading to a curve for overall GI toxicity (R2 = 0.98), could be determined. For these symptoms, mainly rectal thresholds were found throughout low and high doses (10 Gy ≤ equivalent dose in 2-Gy fractions using α/β = 3Gy (EQD23) ≤ 50 Gy and 55 Gy ≤ EQD23 ≤ 78 Gy, respectively). For BT with or without EBRT, dose/volume tolerances were also mainly identified for GI toxicity (GI, 14 thresholds; GU, 4 thresholds; SD, 2 thresholds) with the largest number of DVH thresholds concerning rectal bleeding (5 thresholds).

CONCLUSIONS

Updated dose/volume tolerances after QUANTEC were found for 17 GI, GU, or SD symptoms. A DVH curve described the relationship between dose/volume tolerances across 5 GI symptoms after EBRT alone. Restricting treatments for EBRT alone using the lower boundaries of this curve is likely to limit overall GI toxicity, but this should be explored prospectively. Dose/volume tolerances for GU and SD toxicity after EBRT alone and after BT with or without EBRT were scarce and support further research including data-sharing initiatives to untangle the dose/volume relationships for these symptoms.

Treatment data and technical process challenges for practical big data efforts in radiation oncology

The term Big Data has come to encompass a number of concepts and uses within medicine. This paper lays out the relevance and application of large collections of data in the radiation oncology community. We describe the potential importance and uses in clinical practice. The important concepts are then described and how they have been or could be implemented are discussed. Impediments to progress in the collection and use of sufficient quantities of data are also described. Finally, recommendations for how the community can move forward to achieve the potential of big data in radiation oncology are provided.

Cancer risk assessment in modern radiotherapy workflow with medical big data

Modern radiotherapy (RT) is being enriched by big digital data and intensive technology. Multimodality image registration, intelligence-guided planning, real-time tracking, image-guided RT (IGRT), and automatic follow-up surveys are the products of the digital era. Enormous digital data are created in the process of treatment, including benefits and risks. Generally, decision making in RT tries to balance these two aspects, which is based on the archival and retrieving of data from various platforms. However, modern risk-based analysis shows that many errors that occur in radiation oncology are due to failures in workflow. These errors can lead to imbalance between benefits and risks. In addition, the exact mechanism and dose-response relationship for radiation-induced malignancy are not well understood. The cancer risk in modern RT workflow continues to be a problem. Therefore, in this review, we develop risk assessments based on our current knowledge of IGRT and provide strategies for cancer risk reduction. Artificial intelligence (AI) such as machine learning is also discussed because big data are transforming RT via AI.

U-CAN: a prospective longitudinal collection of biomaterials and clinical information from adult cancer patients in Sweden

BACKGROUND

Progress in cancer biomarker discovery is dependent on access to high-quality biological materials and high-resolution clinical data from the same cases. To overcome current limitations, a systematic prospective longitudinal sampling of multidisciplinary clinical data, blood and tissue from cancer patients was therefore initiated in 2010 by Uppsala and Umeå Universities and involving their corresponding University Hospitals, which are referral centers for one third of the Swedish population.

MATERIAL AND METHODS

Patients with cancer of selected types who are treated at one of the participating hospitals are eligible for inclusion. The healthcare-integrated sampling scheme encompasses clinical data, questionnaires, blood, fresh frozen and formalin-fixed paraffin-embedded tissue specimens, diagnostic slides and radiology bioimaging data.

RESULTS

In this ongoing effort, 12,265 patients with brain tumors, breast cancers, colorectal cancers, gynecological cancers, hematological malignancies, lung cancers, neuroendocrine tumors or prostate cancers have been included until the end of 2016. From the 6914 patients included during the first five years, 98% were sampled for blood at diagnosis, 83% had paraffin-embedded and 58% had fresh frozen tissues collected. For Uppsala County, 55% of all cancer patients were included in the cohort.

CONCLUSIONS

Close collaboration between participating hospitals and universities enabled prospective, longitudinal biobanking of blood and tissues and collection of multidisciplinary clinical data from cancer patients in the U-CAN cohort. Here, we summarize the first five years of operations, present U-CAN as a highly valuable cohort that will contribute to enhanced cancer research and describe the procedures to access samples and data.

Automated customized retrieval of radiotherapy data for clinical trials, audit and research

OBJECTIVE

To enable fast and customizable automated collection of radiotherapy (RT) data from tomotherapy storage.

METHODS

Human-readable data maps (TagMaps) were created to generate DICOM-RT (Digital Imaging and Communications in Medicine standard for Radiation Therapy) data from tomotherapy archives, and provided access to "hidden" information comprising delivery sinograms, positional corrections and adaptive-RT doses.

RESULTS

797 data sets totalling 25,000 scans were batch-exported in 31.5 h. All archived information was restored, including the data not available via commercial software. The exported data were DICOM-compliant and compatible with major commercial tools including RayStation, Pinnacle and ProSoma. The export ran without operator interventions.

CONCLUSION

The TagMap method for DICOM-RT data modelling produced software that was many times faster than the vendor's solution, required minimal operator input and delivered high volumes of vendor-identical DICOM data. The approach is applicable to many clinical and research data processing scenarios and can be adapted to recover DICOM-RT data from other proprietary storage types such as Elekta, Pinnacle or ProSoma. Advances in knowledge: A novel method to translate data from proprietary storage to DICOM-RT is presented. It provides access to the data hidden in electronic archives, offers a working solution to the issues of data migration and vendor lock-in and paves the way for large-scale imaging and radiomics studies.

DBCG hypo trial validation of radiotherapy parameters from a national data bank versus manual reporting

INTRODUCTION

The current study evaluates the data quality achievable using a national data bank for reporting radiotherapy parameters relative to the classical manual reporting method of selected parameters.

METHODS

The data comparison is based on 1522 Danish patients of the DBCG hypo trial with data stored in the Danish national radiotherapy data bank. In line with standard DBCG trial practice selected parameters were also reported manually to the DBCG database. Categorical variables are compared using contingency tables, and comparison of continuous parameters is presented in scatter plots.

RESULTS

For categorical variables 25 differences between the data bank and manual values were located. Of these 23 were related to mistakes in the manual reported value whilst the remaining two were a wrong classification in the data bank. The wrong classification in the data bank was related to lack of dose information, since the two patients had been treated with an electron boost based on a manual calculation, thus data was not exported to the data bank, and this was not detected prior to comparison with the manual data. For a few database fields in the manual data an ambiguity of the parameter definition of the specific field is seen in the data. This was not the case for the data bank, which extract all data consistently.

CONCLUSIONS

In terms of data quality the data bank is superior to manually reported values. However, there is a need to allocate resources for checking the validity of the available data as well as ensuring that all relevant data is present. The data bank contains more detailed information, and thus facilitates research related to the actual dose distribution in the patients.

The big data effort in radiation oncology: Data mining or data farming?

Although large volumes of information are entered into our electronic health care records, radiation oncology information systems and treatment planning systems on a daily basis, the goal of extracting and using this big data has been slow to emerge. Development of strategies to meet this goal is aided by examining issues with a data farming instead of a data mining conceptualization. Using this model, a vision of key data elements, clinical process changes, technology issues and solutions, and role for professional societies is presented. With a better view of technology, process and standardization factors, definition and prioritization of efforts can be more effectively directed.