Artificial Intelligence (AI) models in radiation therapy are being developed with increasing pace. Despite this, the radiation therapy community has not widely adopted these models in clinical practice. A cohesive guideline on how to develop, report and clinically validate AI algorithms might help bridge this gap. A Delphi process with all co-authors was followed to determine which topics should be addressed in this comprehensive guideline. Separate sections of the guideline, including Statements, were written by subgroups of the authors and discussed with the whole group at several meetings. Statements were formulated and scored as highly recommended or recommended. The following topics were found most relevant: Decision making, image analysis, volume segmentation, treatment planning, patient specific quality assurance of treatment delivery, adaptive treatment, outcome prediction, training, validation and testing of AI model parameters, model availability for others to verify, model quality assurance/updates and upgrades, ethics. Key references were given together with an outlook on current hurdles and possibilities to overcome these. 19 Statements were formulated. A cohesive guideline has been written which addresses main topics regarding AI in radiation therapy. It will help to guide development, as well as transparent and consistent reporting and validation of new AI tools and facilitate adoption.
1 November 2020 in International Journal of Radiation Oncology - Biology - Physics (IJROBP)
For lung cancer patients treated with radiotherapy, dose to the heart is associated with excess mortality. However, it is often not feasible to spare the whole heart. Therefore, our aim is to define cardiac sub-structure(s) and dose thresholds which optimally reduce early mortality. 14 cardiac sub-structures were delineated on 5 template patients with representative anatomies. 1,161 non-small cell lung cancer patients were non-rigidly registered to these 5 template anatomies, mapping their radiotherapy dose. Mean and maximum dose to each sub-structure were extracted and the means evaluated as input to prediction models. The cohort was bootstrapped into two variable reduction techniques: elastic-net LASSO and random forest survival model. Each was optimised to extract variables contributing most to overall survival, model coefficients were evaluated to select these sub-structures. The most important variables, common to both models, were selected and evaluated in multivariable cox-proportional hazard models. A threshold dose was defined and Kaplan-Meier survival curves plotted. 978 patients remained after visual QA of the registration. Ranking the model coefficients across the bootstraps selected the maximum dose to the right atrium, right coronary artery and ascending aorta as the most important factors associated with survival. The maximum dose to the combined cardiac region showed significance in the multivariable model, hazard ratio 1.01Gy-1, p=0.03 after accounting for tumour volume (p<0.001), N-stage (p<0.01) and performance status (p=0.01). The optimal threshold for the maximum dose, equivalent dose in 2Gy fractions, was 23Gy. Kaplan-Meier survival curves showed a significant split, log-rank p=0.008. The max dose to the combined cardiac region encompassing the right atrium, right coronary artery and ascending aorta was found to have the greatest impact on patient survival. A maximum EQD2 dose of 23Gy was identified and should be considered as a dose limit in future studies.
The RTOG 0617 trial presented a worse survival for patients with lung cancer treated in the high-dose (74 Gy) arm. In multivariable models, radiation level and whole-heart volumetric dose parameters were associated with survival. In this work, we consider heart subregions to explain the observed survival difference between radiation levels. Voxel-based analysis identified anatomical regions where the dose was associated with survival. Bootstrapping clinical and dosimetric variables into an elastic net model selected variables associated with survival. Multivariable Cox regression survival models assessed the significance of dose to the heart subregion, compared with whole heart v5 and v30. Finally, the trial outcome was assessed after propensity score matching of patients on lung dose, heart subregion dose, and tumor volume. A total of 458 patients were eligible for voxel-based analysis. A region of significance (p < 0.001) was identified in the base of the heart. Bootstrapping selected mean lung dose, radiation level, log tumor volume, and heart region dose. The multivariable Cox model exhibited dose to the heart region (p = 0.02), and tumor volume (p = 0.03) were significantly associated with survival, and radiation level was not significant (p = 0.07). The models exhibited that whole heart v5 and v30 were not associated with survival, with radiation level being significant (p < 0.05). In the matched cohort, no significant survival difference was seen between radiation levels. Dose to the base of the heart is associated with overall survival, partly removing the radiation level effect, and explaining that worse survival in the high-dose arm is owing, in part, to the heart subregion dose. By defining a heart avoidance region, future dose escalation trials may be feasible.