Novel dosimetric validation of a commercial CT scanner based deep learning automated contour solution for prostate radiotherapy.

PURPOSE: OAR delineation accuracy influences: (i) a patient's optimised dose distribution (PD), (ii) the reported doses (RD) presented at approval, which represent plan quality. This study utilised a novel dosimetric validation methodology, comprehensively evaluating a new CT-scanner-based AI contouring solution in terms of PD and RD within an automated planning workflow. METHODS: 20 prostate patients were selected to evaluate AI contouring for rectum, bladder, and proximal femurs. Five planning 'pipelines' were considered; three using AI contours with differing levels of manual editing (nominally none (AIStd), minor editing in specific regions (AIMinEd), and fully corrected (AIFullEd)). Remaining pipelines were manual delineations from two observers (MDOb1, MDOb2). Automated radiotherapy plans were generated for each pipeline. Geometric and dosimetric agreement of contour sets AIStd, AIMinEd, AIFullEd and MDOb2 were evaluated against the reference set MDOb1. Non-inferiority of AI pipelines was assessed, hypothesising that compared to MDOb1, absolute deviations in metrics for AI contouring were no greater than that from MDOb2. RESULTS: Compared to MDOb1, organ delineation time was reduced by 24.9 min (96 %), 21.4 min (79 %) and 12.2 min (45 %) for AIStd, AIMinEd and AIFullEd respectively. All pipelines exhibited generally good dosimetric agreement with MDOb1. For RD, median deviations were within ± 1.8 cm3, ± 1.7 % and ± 0.6 Gy for absolute volume, relative volume and mean dose metrics respectively. For PD, respective values were within ± 0.4 cm3, ± 0.5 % and ± 0.2 Gy. Statistically (p < 0.05), AIMinEd and AIFullEd were dosimetrically non-inferior to MDOb2. CONCLUSIONS: This novel dosimetric validation demonstrated that following targeted minor editing (AIMinEd), AI contours were dosimetrically non-inferior to manual delineations, reducing delineation time by 79 %.

Multi-institutional evaluation of a Pareto navigation guided automated radiotherapy planning solution for prostate cancer.

BACKGROUND: Current automated planning solutions are calibrated using trial and error or machine learning on historical datasets. Neither method allows for the intuitive exploration of differing trade-off options during calibration, which may aid in ensuring automated solutions align with clinical preference. Pareto navigation provides this functionality and offers a potential calibration alternative. The purpose of this study was to validate an automated radiotherapy planning solution with a novel multi-dimensional Pareto navigation calibration interface across two external institutions for prostate cancer. METHODS: The implemented 'Pareto Guided Automated Planning' (PGAP) methodology was developed in RayStation using scripting and consisted of a Pareto navigation calibration interface built upon a 'Protocol Based Automatic Iterative Optimisation' planning framework. 30 previous patients were randomly selected by each institution (IA and IB), 10 for calibration and 20 for validation. Utilising the Pareto navigation interface automated protocols were calibrated to the institutions' clinical preferences. A single automated plan (VMATAuto) was generated for each validation patient with plan quality compared against the previously treated clinical plan (VMATClinical) both quantitatively, using a range of DVH metrics, and qualitatively through blind review at the external institution. RESULTS: PGAP led to marked improvements across the majority of rectal dose metrics, with Dmean reduced by 3.7 Gy and 1.8 Gy for IA and IB respectively (p < 0.001). For bladder, results were mixed with low and intermediate dose metrics reduced for IB but increased for IA. Differences, whilst statistically significant (p < 0.05) were small and not considered clinically relevant. The reduction in rectum dose was not at the expense of PTV coverage (D98% was generally improved with VMATAuto), but was somewhat detrimental to PTV conformality. The prioritisation of rectum over conformality was however aligned with preferences expressed during calibration and was a key driver in both institutions demonstrating a clear preference towards VMATAuto, with 31/40 considered superior to VMATClinical upon blind review. CONCLUSIONS: PGAP enabled intuitive adaptation of automated protocols to an institution's planning aims and yielded plans more congruent with the institution's clinical preference than the locally produced manual clinical plans.

Palliative radiotherapy combined with stent insertion to reduce recurrent dysphagia in oesophageal cancer patients: the ROCS RCT.

BACKGROUND: Most patients with oesophageal cancer present with incurable disease. For those with advanced disease, the mean survival is 3-5 months. Treatment emphasis is therefore on effective palliation, with the majority of patients requiring intervention for dysphagia. Insertion of a self-expanding metal stent provides rapid relief but dysphagia may recur within 3 months owing to tumour progression. Evidence reviews have called for trials of interventions combined with stenting to better maintain the ability to swallow. OBJECTIVES: The Radiotherapy after Oesophageal Cancer Stenting (ROCS) study examined the effectiveness of palliative radiotherapy, combined with insertion of a stent, in maintaining the ability to swallow. The trial also examined the impact that the ability to swallow had on quality of life, bleeding events, survival and cost-effectiveness. DESIGN: A pragmatic, multicentre, randomised controlled trial with follow-up every 4 weeks for 12 months. An embedded qualitative study examined trial experiences in a participant subgroup. SETTING: Participants were recruited in secondary care, with all planned follow-up at home. PARTICIPANTS: Patients who were referred for stent insertion as the primary management of dysphagia related to incurable oesophageal cancer. INTERVENTIONS: Following stent insertion, the external beam radiotherapy arm received palliative oesophageal radiotherapy at a dose of 20 Gy in five fractions or 30 Gy in 10 fractions. MAIN OUTCOME MEASURES: The primary outcome was the difference in the proportion of participants with recurrent dysphagia, or death, at 12 weeks. Recurrent dysphagia was defined as deterioration of ≥ 11 points on the dysphagia scale of the European Organisation of Research and Treatment of Cancer Quality of Life Questionnaire oesophago-gastric module questionnaire. Secondary outcomes included quality of life, bleeding risk and survival. RESULTS: The study recruited 220 patients: 112 were randomised to the usual-care arm and 108 were randomised to the external beam radiotherapy arm. There was no evidence that radiotherapy reduced recurrence of dysphagia at 12 weeks (48.6% in the usual-care arm compared with 45.3% in the external beam radiotherapy arm; adjusted odds ratio 0.82, 95% confidence interval 0.40 to 1.68; p = 0.587) and it was less cost-effective than stent insertion alone. There was no difference in median survival or key quality-of-life outcomes. There were fewer bleeding events in the external beam radiotherapy arm. Exploration of patient experience prompted changes to trial processes. Participants in both trial arms experienced difficulty in managing the physical and psychosocial aspects of eating restriction and uncertainties of living with advanced oesophageal cancer. LIMITATIONS: Change in timing of the primary outcome to 12 weeks may affect the ability to detect a true intervention effect. However, consistency of results across sensitivity analyses is robust, including secondary analysis of dysphagia deterioration-free survival. CONCLUSIONS: Widely accessible palliative external beam radiotherapy in combination with stent insertion does not reduce the risk of dysphagia recurrence at 12 weeks, does not have an impact on survival and is less cost-effective than inserting a stent alone. Reductions in bleeding events should be considered in the context of patient-described trade-offs of fatigue and burdens of attending hospital. Trial design elements including at-home data capture, regular multicentre nurse meetings and qualitative enquiry improved recruitment/data capture, and should be considered for future studies. FUTURE WORK: Further studies are required to identify interventions that improve stent efficacy and to address the multidimensional challenges of eating and nutrition in this patient population. TRIAL REGISTRATION: Current Controlled Trials ISRCTN12376468 and Clinicaltrials.gov NCT01915693. FUNDING: This project was funded by the National Institute for Health Research (NIHR) Health Technology Assessment programme and will be published in full in Health Technology Assessment; Vol. 25, No. 31. See the NIHR Journals Library website for further project information.

Most people are diagnosed with oesophageal (gullet) cancer when it is already at an advanced stage. Losing the ability to swallow food and even fluids is very common when patients are approaching the last months of life. Placing a flexible metal tube, or stent, in the gullet opens it up and improves the ability to swallow quickly. Unfortunately this can fail after around 3 months because the cancer grows and presses on the stent. We designed this trial to see if giving a small dose of radiotherapy alongside insertion of the stent would allow more people to remain swallowing well after 3 months. This could then improve their quality of life and reduce hospitalisation towards the end of life. It may also reduce bleeding from the gullet, as well as other symptoms. We recruited 220 people across the UK, randomly assigning them to have the stent as usual or the stent and a low dose of radiotherapy. We collected a lot of information from the participants at home on how the cancer, the stent and the radiotherapy affected their ability to swallow and their quality of life. Overall, the study showed that the radiotherapy did not improve the ability to swallow 3 months following stent insertion and was less cost-effective than stent insertion alone. It seemed to reduce the risk of bleeding from the tumour itself, but patients found that radiotherapy made them tired and attending extra hospital visits could be troublesome. We also learned that, even after a stent was inserted, patients still struggled with food and needed more support with managing daily life with the stent. The trial results are important. They show that, to answer questions such as these, studies should use different ways of assessing what works, particularly focusing on patients' and families' viewpoints. The results will guide doctors to not routinely give radiotherapy in this situation. The results also suggest that, after the insertion of a stent, patients need extra help in managing their diet, their worries about the stent and their worries about the future.

Palliative radiotherapy after oesophageal cancer stenting (ROCS): a multicentre, open-label, phase 3 randomised controlled trial.

BACKGROUND: Patients with advanced oesophageal cancer have a median survival of 3-6 months, and most require intervention for dysphagia. Self-expanding metal stent (SEMS) insertion is the most typical form of palliation in these patients, but dysphagia deterioration and re-intervention are common. This study examined the efficacy of adjuvant external beam radiotherapy (EBRT) compared with usual care alone in preventing dysphagia deterioration and reducing service use after SEMS insertion. METHODS: This was a multicentre, open-label, phase 3 randomised controlled trial based at cancer centres and acute care hospitals in England, Scotland, and Wales. Patients (aged ≥16 years) with incurable oesophageal carcinoma receiving stent insertion for primary management of dysphagia were randomly assigned (1:1) to receive usual care alone or EBRT (20 Gy in five fractions or 30 Gy in ten fractions) plus usual care after stent insertion. Usual care was implemented according to need as identified by the local multidisciplinary team (MDT). Randomisation was via the method of minimisation stratified by treating centre, stage at diagnosis (I-III vs IV), histology (squamous or non-squamous), and MDT intent to give chemotherapy (yes vs no). The primary outcome was difference in proportions of participants with dysphagia deterioration (>11 point decrease on patient-reported European Organisation for Research and Treatment of Cancer quality of life questionnaire-oesophagogastric module [QLQ-OG25], or a dysphagia-related event consistent with such a deterioration) or death by 12 weeks in a modified intention-to-treat (ITT) population, which excluded patients who did not have a stent inserted and those without a baseline QLQ-OG25 assessment. Secondary outcomes included survival, quality of life (QoL), morbidities (including time to first bleeding event or hospital admission for bleeding event and first dysphagia-related stent complications or re-intervention), and cost-effectiveness. Safety analysis was undertaken in the modified ITT population. The study is registered with the International Standard Randomised Controlled Trial registry, ISRCTN12376468, and ClinicalTrials.gov, NCT01915693, and is completed. FINDINGS: 220 patients were randomly assigned between Dec 16, 2013, and Aug 24, 2018, from 23 UK centres. The modified ITT population (n=199) comprised 102 patients in the usual care group and 97 patients in the EBRT group. Radiotherapy did not reduce dysphagia deterioration, which was reported in 36 (49%) of 74 patients receiving usual care versus 34 (45%) of 75 receiving EBRT (adjusted odds ratio 0·82 [95% CI 0·40-1·68], p=0·59) in those with complete data for the primary endpoint. No significant difference was observed in overall survival: median overall survival was 19·7 weeks (95% CI 14·4-27·7) with usual care and 18·9 weeks (14·7-25·6) with EBRT (adjusted hazard ratio 1·06 [95% CI 0·78-1·45], p=0·70; n=199). Median time to first bleeding event or hospital admission for a bleeding event was 49·0 weeks (95% CI 33·3-not reached) with usual care versus 65·9 weeks (52·7-not reached) with EBRT (adjusted subhazard ratio 0·52 [95% CI 0·28-0·97], p=0·038; n=199). No time versus treatment interaction was observed for prespecified QoL outcomes. We found no evidence of differences between trial group in time to first stent complication or re-intervention event. The most common (grade 3-4) adverse event was fatigue, reported in 19 (19%) of 102 patients receiving usual care alone and 22 (23%) of 97 receiving EBRT. On cost-utility analysis, EBRT was more expensive and less efficacious than usual care. INTERPRETATION: Patients with advanced oesophageal cancer having SEMS insertion for the primary management of their dysphagia did not gain additional benefit from concurrent palliative radiotherapy and it should not be routinely offered. For a minority of patients clinically considered to be at high risk of tumour bleeding, concurrent palliative radiotherapy might reduce bleeding risk and the need for associated interventions. FUNDING: National Institute for Health Research Health Technology Assessment Programme.

Evaluating the application of Pareto navigation guided automated radiotherapy treatment planning to prostate cancer.

BACKGROUND AND PURPOSE: Current automated planning methods do not allow for the intuitive exploration of clinical trade-offs during calibration. Recently a novel automated planning solution, which is calibrated using Pareto navigation principles, has been developed to address this issue. The purpose of this work was to clinically validate the solution for prostate cancer patients with and without elective nodal irradiation. MATERIALS AND METHODS: For 40 randomly selected patients (20 prostate and seminal vesicles (PSV) and 20 prostate and pelvic nodes (PPN)) automatically generated volumetric modulated arc therapy plans (VMATAuto) were compared against plans created by expert dosimetrists under clinical conditions (VMATClinical) and no time pressures (VMATIdeal). Plans were compared through quantitative comparison of dosimetric parameters and blind review by an oncologist. RESULTS: Upon blind review 39/40 and 33/40 VMATAuto plans were considered preferable or equal to VMATClinical and VMATIdeal respectively, with all deemed clinically acceptable. Dosimetrically, VMATAuto, VMATClinical and VMATIdeal were similar, with observed differences generally of low clinical significance. Compared to VMATClinical, VMATAuto reduced hands-on planning time by 94% and 79% for PSV and PPN respectively. Total planning time was significantly reduced from 22.2 mins to 14.0 mins for PSV, with no significant reduction observed for PPN. CONCLUSIONS: A novel automated planning solution has been evaluated, whose Pareto navigation based calibration enabled clinical decision-making on trade-off balancing to be intuitively incorporated into automated protocols. It was successfully applied to two sites of differing complexity and robustly generated high quality plans in an efficient manner.

Utilisation of Pareto navigation techniques to calibrate a fully automated radiotherapy treatment planning solution.

BACKGROUND AND PURPOSE: Current automated radiotherapy planning solutions do not allow for the intuitive exploration of different treatment options during protocol calibration. This work introduces an automated planning solution, which aims to address this problem through incorporating Pareto navigation techniques into the calibration process. MATERIALS AND METHODS: For each tumour site a set of planning goals is defined. Utilising Pareto navigation techniques an operator calibrates the solution through intuitively exploring different treatment options: selecting the optimum balancing of competing planning goals for the given site. Once calibrated, fully automated plan generation is possible, with specific algorithms implemented to ensure trade-off balancing of new patients is consistent with that during calibration. Using the proposed methodology the system was calibrated for prostate and seminal vesicle treatments. The resultant solution was validated through quantitatively comparing the dose distribution of automatically generated plans (VMATAuto) against the previous clinical plan, for ten randomly selected patients. RESULTS: VMATAuto yielded statistically significant improvements in: PTV conformity indices, high dose bladder metrics, mean bowel dose, and the majority of rectum dose metrics. Of particular note was the reduction in mean rectum dose (median 25.1 Gy vs. 27.5 Gy), rectum V24.3Gy (median 41.1% vs. 46.4%), and improvement in the conformity index for the primary PTV (median 0.86 vs. 0.79). Dosimetric improvements were not at the cost of other dose metrics. CONCLUSIONS: An automated planning methodology with a Pareto navigation based calibration has been developed, which enables the complex balancing of competing trade-offs to be intuitively incorporated into automated protocols.

A comparison of electronic portal dosimetry verification methods for use in stereotactic radiotherapy.

Three methods of transit dosimetry using Electronic Portal Imaging Devices (EPIDs) were investigated for use in routine in-vivo dosimetry for cranial stereotactic radiosurgery and radiotherapy. The approaches examined were (a) A full Monte Carlo (MC) simulation of radiation transport through the linear accelerator and patient; (b) Calculation of the expected fluence by a treatment planning system (TPS); (c) Point doses calculated along the central axis compared to doses calculated using parameters acquired using the EPID. A dosimetric comparison of each of the three methods predicted doses at the imager plane to within ±5% and a gamma comparison for the MC and TPS based approaches showed good agreement for a range of dose and distance to agreement criteria. The MC technique was most time consuming, followed by the TPS calculation with the point dose calculation significantly quicker than the other methods.

The effect of planning algorithms in oesophageal radiotherapy in the context of the SCOPE 1 trial.

BACKGROUND AND PURPOSE: In radiotherapy clinical trials multiple centres contribute to patient recruitment. Depending on the calculation algorithm used, the reported dose distributions may differ significantly: broadly, the results for algorithms which do not model lateral electron transport (type a) give less accurate results than the more recently available algorithms that do (type b) when compared to Monte Carlo simulations and measurements. Clinical implementation studies for type b algorithms have yet to be reported for oesophageal radiotherapy. Furthermore, clinical trials must ensure an equivalent effect of the treatment regardless of calculation method. This retrospective planning study aims to define guidance for type b planning in a UK oesophageal clinical trial, to enable acceptable consistency of dose distributions regardless of algorithm, and allow for the improved calculation accuracy of type b to be incorporated into the optimization. MATERIALS AND METHODS: Fifteen patient data sets were planned using a single type a algorithm. Plans were recalculated using a single type b algorithm and subsequently re-optimized with the type b in accordance with optimization rules. The changes in absolute dose at the point of prescription for type a were compared to the recalculated type b. Dose-volume data for organs at risk (OARs), and target volumes were compared, and the volume of the planning target volume (PTV) receiving 95% of the prescribed dose (V95%) was compared to the percentage of PTV overlapping with lung. RESULTS: Dose at the prescription point decreased by 0.69% on average (SD=0.71), p=0.0021, for type b compared to that for type a. For the re-optimized type b, the OAR doses corresponding to the trial dose-volume constraints were maintained within 1.0% of the type a levels on average. Reductions in the mean PTV V95% of 9.3% and 3.8% were observed for the recalculated and re-optimized type b plans, respectively, when compared to the mean PTV V95% for type a. For the re-optimized type b there is a correlation between PTV V95% and the percentage of PTV overlapping lung (R(2)=0.4979). CONCLUSIONS: Plan optimization with the type b algorithm results in improved PTV V95%. Using our suggested optimization rules, equivalent OAR doses can be maintained with both types. For type b, this requires a measured level of compromise to PTV in low density tissue, quantified by the relationship between PTV V95% and the percentage of PTV in lung.