[Database supported electronic retrospective analyses in radiation oncology: establishing a workflow using the example of pancreatic cancer]

Automated data mining of a plan-check database and example application

Purpose

The aim of this work was to present the development and example application of an automated data mining software platform that preforms bulk analysis of results and patient data passing through the 3D plan and delivery QA system, Mobius3D.

Methods

Python, matlab, and Java were used to create an interface that reads JavaScript Object Notation (JSON) created for every approved Mobius3D pre‐treatment plan‐check. The aforementioned JSON files contain all the information for every pre‐treatment QA check performed by Mobius3D, including all 3D dose, CT, structure set information, as well as all plan information and patient demographics. Two Graphical User Interfaces (GUIs) were created, the first is called Mobius3D‐Database (M3D‐DB) and presents the check results in both filterable tabular and graphical form. These data are presented for all patients and includes mean dose differences, 90% coverage, 3D gamma pass rate percentages, treatment sites, machine, beam energy, Multi‐Leaf Collimator (MLC) mode, treatment planning system (TPS), plan names, approvers, dates and times. Group statistics and statistical process control levels are then calculated based on filter settings. The second GUI, called Mobius3D organ at risk (M3DOAR), analyzes dose‐volume histogram data for all patients and all Organs‐at‐Risk (OAR). The design of the software is such that all treatment parameters and treatment site information are able to be filtered and sorted with the results, plots, and statistics updated.

Results

The M3D‐DB software can summarize and filter large numbers of plan‐checks from Mobius3D. The M3DOAR software is also able to analyze large amounts of dose‐volume data for patient groups which may prove useful in clinical trials, where OAR doses for large numbers of patients can be compared and correlated. Target DVHs can also be analyzed en mass.

Conclusions

This work demonstrates a method to extract the large amount of treatment data for every patient that is stored by Mobius3D but not easily accessible. With scripting, it is possible to mine this data for research and clinical trials as well as patient and TPS QA.

Electronic Support for Retrospective Analysis in the Field of Radiation Oncology: Proof of Principle Using an Example of Fractionated Stereotactic Radiotherapy of 251 Meningioma Patients

Introduction

The purpose of this study is to verify the possible benefit of a clinical data warehouse (DWH) for retrospective analysis in the field of radiation oncology.

Material and methods

We manually and electronically (using DWH) evaluated demographic, radiotherapy, and outcome data from 251 meningioma patients, who were irradiated from January 2002 to January 2015 at the Department of Radiation Oncology of the Erlangen University Hospital. Furthermore, we linked the Oncology Information System (OIS) MOSAIQ^® to the DWH in order to gain access to irradiation data. We compared the manual and electronic data retrieval method in terms of congruence of data, corresponding time, and personal requirements (physician, physicist, scientific associate).

Results

The electronically supported data retrieval (DWH) showed an average of 93.9% correct data and significantly (p = 0.009) better result compared to manual data retrieval (91.2%). Utilizing a DWH enables the user to replace large amounts of manual activities (668 h), offers the ability to significantly reduce data collection time and labor demand (35 h), while simultaneously improving data quality. In our case, work time for manually data retrieval was 637 h for the scientific assistant, 26 h for the medical physicist, and 5 h for the physician (total 668 h).

Conclusion

Our study shows that a DWH is particularly useful for retrospective analysis in the radiation oncology field. Routine clinical data for a large patient group can be provided ready for analysis to the scientist and data collection time can be significantly reduced. Furthermore, linking multiple data sources in a DWH offers the ability to improve data quality for retrospective analysis, and future research can be simplified.

Review of Developments in Electronic, Clinical Data Collection, and Documentation Systems over the Last Decade - Are We Ready for Big Data in Routine Health Care?

Recently, information availability has become more elaborate and widespread, and treatment decisions are based on a multitude of factors, including imaging, molecular or pathological markers, surgical results, and patient’s preference. In this context, the term “Big Data” evolved also in health care. The “hype” is heavily discussed in literature. In interdisciplinary medical specialties, such as radiation oncology, not only heterogeneous and voluminous amount of data must be evaluated but also spread in different styles across various information systems. Exactly this problem is also referred to in many ongoing discussions about Big Data – the “three V’s”: volume, velocity, and variety. We reviewed 895 articles extracted from the NCBI databases about current developments in electronic clinical data management systems and their further analysis or postprocessing procedures. Few articles show first ideas and ways to immediately make use of collected data, particularly imaging data. Many developments can be noticed in the field of clinical trial or analysis documentation, mobile devices for documentation, and genomics research. Using Big Data to advance medical research is definitely on the rise. Health care is perhaps the most comprehensive, important, and economically viable field of application.

Data management, documentation and analysis systems in radiation oncology: a multi-institutional survey

Introduction

Recently, information availability has become more elaborate and widespread, and treatment decisions are based on a multitude of factors. Gathering relevant data, also referred to as Big Data, is therefore critical for reaching the best patient care, and enhancing interdisciplinary and clinical research. Combining patient data from all involved systems is essential to prepare unstructured data for analyses. This demands special coordination in data management. Our study aims to characterize current developments in German-speaking hospital departments and practices. We successfully conducted the survey with the members of the Deutsche Gesellschaft für Radioonkologie (DEGRO).

Methods

A questionnaire was developed consisting of 17 questions related to data management, documentation and clinical trial analyses, reflecting the clinical topics such as basic patient information, imaging, follow-up information as well as connection of documentation tools with radiooncological treatment planning machines.

Results

A total of 44 institutions completed the online survey (University hospitals n = 17, hospitals n = 13, practices/institutes n = 14). University hospitals, community hospitals and private practices are equally equipped concerning IT infrastructure for clinical use. However, private practices have a low interest in research work. All respondents stated the biggest obstacles about introducing a documentation system into their unit lie in funding and support of the central IT departments. Only 27 % (12/44) of responsible persons are specialists for documentation and data management.

Conclusion

Our study gives an understanding of the challenges and solutions we need to be looking at for medical data storage. In the future, inter-departmental cross-links will enable the radiation oncology community to generate large-scale analyses.

Electronic supplementary material

The online version of this article (doi:10.1186/s13014-015-0543-0) contains supplementary material, which is available to authorized users.

Changes in Gross Tumor Volume and Organ Motion Analysis During Neoadjuvant Radiochemotherapy in Patients With Locally Advanced Pancreatic Cancer Using an In-House Analysis System

BACKGROUND AND PURPOSE

During radiation treatment, movement of the target and organs at risks as well as tumor response can significantly influence dose distribution. This is highly relevant in patients with pancreatic cancer, where organs at risk lie in close proximity to the target.

MATERIAL AND METHODS

Data sets of 10 patients with locally advanced pancreatic cancer were evaluated. Gross tumor volume deformation was analyzed. Dose changes to organs at risk were determined with focus on kidneys both without adaptive radiotherapy compensation and with replanning based on weekly acquired computed tomography scans.

RESULTS

During irradiation, gross tumor volume changes between 0% and 26% and moves within a radius of 5 to 16 mm. Required maximal dose to organs at risk for kidneys can be met with the current practice of matching computed tomography scans during treatment and adjusting patient position accordingly. Comparison of the mean doses and V15, V20 volumes demonstrated that weekly replanning could bring a significant dose sparing of the left kidney.

CONCLUSION

Manual matching with focus on bony structures can lead to overall acceptable positioning of patients during treatment. Thus, tolerance doses of organs at risk, such as the kidneys, can be met. With adequate margins, normal tissue constraints to organs at risk can be kept as well. Adaptive radiotherapy approaches (in this case with weekly rescanning) reduced dose to organs at risk, which may be especially important for hypofractionated approaches.

Five-year experience with setup and implementation of an integrated database system for clinical documentation and research

In radiation oncology, where treatment concepts are elaborated in interdisciplinary collaborations, handling distributed, large heterogeneous amounts of data efficiently is very important, yet challenging, for an optimal treatment of the patient as well as for research itself. This becomes a strong focus, as we step into the era of modern personalized medicine, relying on various quantitative data information, thus involving the active contribution of multiple medical specialties. Hence, combining patient data from all involved information systems is inevitable for analyses. Therefore, we introduced a documentation and data management system integrated in the clinical environment for electronic data capture. We discuss our concept and five-year experience of a precise electronic documentation system, with special focus on the challenges we encountered. We specify how such a system can be designed and implemented to plan, tailor and conduct (multicenter) clinical trials, ultimately reaching the best clinical performance, and enhancing interdisciplinary and clinical research.

Intensity modulated radiotherapy as neoadjuvant chemoradiation for the treatment of patients with locally advanced pancreatic cancer. Outcome analysis and comparison with a 3D-treated patient cohort

BACKGROUND

To evaluate outcome after intensity modulated radiotherapy (IMRT) compared to 3D conformal radiotherapy (3D-RT) as neoadjuvant treatment in patients with locally advanced pancreatic cancer (LAPC).

MATERIALS AND METHODS

In total, 57 patients with LAPC were treated with IMRT and chemotherapy. A median total dose of 45 Gy to the PTV_baseplan and 54 Gy to the PTV_boost in single doses of 1.8 Gy for the PTV_baseplan and median single doses of 2.2 Gy in the PTV_boost were applied. Outcomes were evaluated and compared to a large cohort of patients treated with 3D-RT.

RESULTS

Overall treatment was well tolerated in all patients and IMRT could be completed without interruptions. Median overall survival was 11 months (range 5-37.5 months). Actuarial overall survival at 12 and 24 months was 36 % and 8 %, respectively. A significant impact on overall survival could only be observed for a decrease in CA 19-9 during treatment, patients with less pre-treatment CA 19-9 than the median, as well as weight loss during treatment. Local progression-free survival was 79 % after 6 months, 39 % after 12 months, and 13 % after 24 months. No factors significantly influencing local progression-free survival could be identified. There was no difference in overall and progression-free survival between 3D-RT and IMRT. Secondary resectability was similar in both groups (26 % vs. 28 %). Toxicity was comparable and consisted mainly of hematological toxicity due to chemotherapy.

CONCLUSION

IMRT leads to a comparable outcome compared to 3D-RT in patients with LAPC. In the future, the improved dose distribution, as well as advances in image-guided radiotherapy (IGRT) techniques, may improve the use of IMRT in local dose escalation strategies to potentially improve outcome.

Development and validation of automatic tools for interactive recurrence analysis in radiation therapy: optimization of treatment algorithms for locally advanced pancreatic cancer

BACKGROUND

In radiation oncology recurrence analysis is an important part in the evaluation process and clinical quality assurance of treatment concepts. With the example of 9 patients with locally advanced pancreatic cancer we developed and validated interactive analysis tools to support the evaluation workflow.

METHODS

After an automatic registration of the radiation planning CTs with the follow-up images, the recurrence volumes are segmented manually. Based on these volumes the DVH (dose volume histogram) statistic is calculated, followed by the determination of the dose applied to the region of recurrence and the distance between the boost and recurrence volume. We calculated the percentage of the recurrence volume within the 80%-isodose volume and compared it to the location of the recurrence within the boost volume, boost + 1 cm, boost + 1.5 cm and boost + 2 cm volumes.

RESULTS

Recurrence analysis of 9 patients demonstrated that all recurrences except one occurred within the defined GTV/boost volume; one recurrence developed beyond the field border/outfield. With the defined distance volumes in relation to the recurrences, we could show that 7 recurrent lesions were within the 2 cm radius of the primary tumor. Two large recurrences extended beyond the 2 cm, however, this might be due to very rapid growth and/or late detection of the tumor progression.

CONCLUSION

The main goal of using automatic analysis tools is to reduce time and effort conducting clinical analyses. We showed a first approach and use of a semi-automated workflow for recurrence analysis, which will be continuously optimized. In conclusion, despite the limitations of the automatic calculations we contributed to in-house optimization of subsequent study concepts based on an improved and validated target volume definition.

Comparison of four target volume definitions for pancreatic cancer. Guidelines for treatment of the lymphatics and the primary tumor

BACKGROUND AND PURPOSE

Target volume definitions for radiotherapy in pancreatic ductal adenocarcinoma (PDAC) vary substantially. Some groups aim to treat the primary tumor only, whereas others include elective lymph nodes (eLNs). eLNs close to the primary tumor are often included unintentionally within the treatment volume, depending on the respective treatment philosophies. We aimed to measure the percentages of anatomical coverage of eLNs by comparing four different contouring guidelines.

PATIENTS AND METHODS

Planning target volumes (PTVs) were contoured using planning computed tomography (CT) scans of 11 patients with PDAC based on the Oxford, RTOG (Radiation Therapy Oncology Group), Michigan, and SCALOP (Selective Chemoradiation in Advanced Localised Pancreatic Cancer trial) guidelines. Clinical target volumes (CTVs) included the peripancreatic, para-aortic, paracaval, celiac trunk, superior mesenteric, and portal vein lymph node areas. Volumetric comparisons of the coverage of all eLN regions were conducted to illustrate the differences between the four contouring strategies.

RESULTS

The PTV sizes of the RTOG and Oxford guidelines were comparable. The SCALOP and Michigan PTV sizes were similar to each other and significantly smaller than the RTOG and Oxford PTVs. A large variability of eLN coverage was found for the various subregions according to the respective contouring strategies.

CONCLUSION

This is the first study to directly compare the percentage of anatomical coverage of eLNs according to four PTVs in the same patient cohort. Potential practical consequences are discussed in detail.