Multi-centre evaluation of variation in cumulative dose assessment in reirradiation scenarios.

BACKGROUND AND PURPOSE: Safe reirradiation relies on assessment of cumulative doses to organs at risk (OARs) across multiple treatments. Different clinical pathways can result in inconsistent estimates. Here, we quantified the consistency of cumulative dose to OARs across multi-centre clinical pathways. MATERIAL AND METHODS: We provided DICOM planning CT, structures and doses for two reirradiation cases: head & neck (HN) and lung. Participants followed their standard pathway to assess the cumulative physical and EQD2 doses (with provided α/ß values), and submitted DVH metrics and a description of their pathways. Participants could also submit physical dose distributions from Course 1 mapped onto the CT of Course 2 using their best available tools. To assess isolated impact of image registrations, a single observer accumulated each submitted spatially mapped physical dose for every participating centre. RESULTS: Cumulative dose assessment was performed by 24 participants. Pathways included rigid (n = 15), or deformable (n = 5) image registration-based 3D dose summation, visual inspection of isodose line contours (n = 1), or summation of dose metrics extracted from each course (n = 3). Largest variations were observed in near-maximum cumulative doses (25.4 - 41.8 Gy for HN, 2.4 - 33.8 Gy for lung OARs), with lower variations in volume/dose metrics to large organs. A standardised process involving spatial mapping of the first course dose to the second course CT followed by summation improved consistency for most near-maximum dose metrics in both cases. CONCLUSION: Large variations highlight the uncertainty in reporting cumulative doses in reirradiation scenarios, with implications for outcome analysis and understanding of published doses. Using a standardised workflow potentially including spatially mapped doses improves consistency in determination of accumulated dose in reirradiation scenarios.

Convolutional neural network and transfer learning for dose volume histogram prediction for prostate cancer radiotherapy.

To adopt a transfer learning approach and establish a convolutional neural network (CNN) model for the prediction of rectum and bladder dose-volume histograms (DVH) in prostate patients treated with a VMAT technique. One hundred forty-four VMAT patients with intermediate or high-risk prostate cancer were included in this study. Data were split into two sets: 120 and 24 patients, respectively. The second set was used for final validation. To ensure the accuracy of the training data, we developed a ground-truth analysis for detecting and correcting for all potential outliers. We used transfer learning in combination with a pre-trained VGG-16 network. We dropped the fully connected layers from the VGG-16 and added a new fully connected neural network. The inputs for the CNN were a 2D image of the volumes contoured in the CT, but we only retained the geometrical information of every CT-slice. The outputs were the corresponding rectum and bladder DVH for every slice. We used a confusion matrix to analyze the performance of our model. Our model achieved 100% and 81% of true positive and true negative predictions, respectively. We have an overall accuracy of 87.5%, a misclassification rate of 12.5%, and a precision of 100%. We have successfully developed a model for reliable prediction of rectum and bladder DVH in prostate patients by applying a previously pre-trained CNN. To our knowledge, this is the first attempt to apply transfer learning to the prediction of DVHs that accounts for the ground truth problem.

Publication of interventional phase 3 and 4 clinical trials in radiation oncology: an observational study.

OBJECTIVES: Clinical trials produce the best data available for decision-making in modern evidence-based medicine. We aimed to determine the rate of non-publication of interventional phase 3 and 4 clinical trials involving patients with cancer undergoing radiotherapy. SETTING: The ClinicalTrials.gov database was searched for interventional phase 3 and 4 trials in radiotherapy with a primary completion date before 1 January 2013. We determined how many of these registry entries have not published the compulsory deposition of their results in the database and performed a systematic search for published studies in peer-reviewed journals. RESULTS: Of 576 trials, 484 (84.0%) did not deposit a summary result in the registry. In addition, 44.9% of them did not publish their results in a peer-reviewed journal. Similar percentages were found for most cancer subtypes: brain (41%), breast (38%), cervical (66%), colorectal (38%), lung (48%), prostate (45%), bladder (56%), head and neck (56%) and lymphoma (33%). CONCLUSION: Our results show that most trials in radiation oncology did not report the results in the registry. Almost half of these trials have not been published in the biomedical literature. This means that a large number of study participants were exposed to the risks of trial participation without the supposed benefits that sharing and publishing of results would offer to future generations of patients.