Analysis of EORTC-1219-DAHANCA-29 trial plans demonstrates the potential of knowledge-based planning to provide patient-specific treatment plan quality assurance

Dose distribution prediction for head-and-neck cancer radiotherapy using a generative adversarial network: influence of input data

Purpose

A three-dimensional deep generative adversarial network (GAN) was used to predict dose distributions for locally advanced head and neck cancer radiotherapy. Given the labor- and time-intensive nature of manual planning target volume (PTV) and organ-at-risk (OAR) segmentation, we investigated whether dose distributions could be predicted without the need for fully segmented datasets.

Materials and methods

GANs were trained/validated/tested using 320/30/35 previously segmented CT datasets and treatment plans. The following input combinations were used to train and test the models: CT-scan only (C); CT+PTVboost/elective (CP); CT+PTVs+OARs+body structure (CPOB); PTVs+OARs+body structure (POB); PTVs+body structure (PB). Mean absolute errors (MAEs) for the predicted dose distribution and mean doses to individual OARs (individual salivary glands, individual swallowing structures) were analyzed.

Results

For the five models listed, MAEs were 7.3 Gy, 3.5 Gy, 3.4 Gy, 3.4 Gy, and 3.5 Gy, respectively, without significant differences among CP-CPOB, CP-POB, CP-PB, among CPOB-POB. Dose volume histograms showed that all four models that included PTV contours predicted dose distributions that had a high level of agreement with clinical treatment plans. The best model CPOB and the worst model PB (except model C) predicted mean dose to within ±3 Gy of the clinical dose, for 82.6%/88.6%/82.9% and 71.4%/67.1%/72.2% of all OARs, parotid glands (PG), and submandibular glands (SMG), respectively. The R2 values (0.17/0.96/0.97/0.95/0.95) of OAR mean doses for each model also indicated that except for model C, the predictions correlated highly with the clinical dose distributions. Interestingly model C could reasonably predict the dose in eight patients, but on average, it performed inadequately.

Conclusion

We demonstrated the influence of the CT scan, and PTV and OAR contours on dose prediction. Model CP was not statistically different from model CPOB and represents the minimum data statistically required to adequately predict the clinical dose distribution in a group of patients.

Implementation, Dosimetric Assessment, and Treatment Validation of Knowledge-Based Planning (KBP) Models in VMAT Head and Neck Radiation Oncology

The aim of this study was to evaluate knowledge-based treatment planning (KBP) models in terms of their dosimetry and deliverability and to investigate their clinical benefits. Three H&N KBP models were built utilizing RapidPlan™, based on the dose prescription, which is given according to the planning target volume (PTV). The training set for each model consisted of 43 clinically acceptable volumetric modulated arc therapy (VMAT) plans. Model quality was assessed and compared to the delivered treatment plans using the homogeneity index (HI), conformity index (CI), structure dose difference (PTV, organ at risk—OAR), monitor units, MU factor, and complexity index. Model deliverability was assessed through a patient-specific quality assurance (PSQA) gamma index-based analysis. The dosimetric assessment showed better OAR sparing for the RapidPlan™ plans and for the low- and high-risk PTV, and the HI, and CI were comparable between the clinical and RapidPlan™ plans, while for the intermediate-risk PTV, CI was better for clinical plans. The 2D gamma passing rates for RapidPlan™ plans were similar or better than the clinical ones using the 3%/3 mm gamma-index criterion. Monitor units, the MU factors, and complexity indices were found to be comparable between RapidPlan™ and the clinical plans. Knowledge-based treatment plans can be safely adapted into clinical routines, providing improved plan quality in a time efficient way while minimizing user variability.

Knowledge-Based Quality Assurance and Model Maintenance in Lung Cancer Radiation Therapy in a Statewide Quality Consortium of Academic and Community Practice Centers

PURPOSE

Locally advanced lung cancer (LALC) treatment planning is often complex due to challenging tradeoffs related to large targets near organs at risk, making the judgment of plan quality difficult. The purpose of this work was to update and maintain a multi-institutional knowledge-based planning (KBP) model developed by a statewide consortium of academic and community practices for use as a plan quality assurance (QA) tool.

METHODS AND MATERIALS

Sixty LALC volumetric-modulated arc therapy plans from 2021 were collected from 24 institutions. Plan quality was scored, with high-quality clinical (HQC) plans selected to update a KBP model originally developed in 2017. The model was validated via automated KBP planning, with 20 cases excluded from the model. Differences in dose-volume histogram metrics in the clinical plans, 2017 KBP model plans, and 2022 KBP model plans were compared. Twenty recent clinical cases not meeting consortium quality metrics were replanned with the 2022 model to investigate potential plan quality improvements.

RESULTS

Forty-seven plans were included in the final KBP model. Compared with the clinical plans, the 2022 model validation plans improved 60%, 65%, and 65% of the lung V20Gy, mean heart dose, and spinal canal D0.03cc metrics, respectively. The 2022 model showed improvements from the 2017 model in hot spot management at the cost of greater lung doses. Of the 20 recent cases not meeting quality metrics, 40% of the KBP model-replanned cases resulted in acceptable plans, suggesting potential clinical plan improvements.

CONCLUSIONS

A multi-institutional KBP model was updated using plans from a statewide consortium. Multidisciplinary plan review resulted in HQC model training plans and model validation resulted in acceptable quality plans. The model proved to be effective at identifying potential plan quality improvements. Work is ongoing to develop web-based training plan review tools and vendor-agnostic platforms to provide the model as a QA tool statewide.

Fast, automated knowledge-based treatment planning for selecting patients for proton therapy based on normal tissue complication probabilities

Purpose

Selecting patients who will benefit from proton therapy is laborious and subjective. We demonstrate a novel automated solution for creating high-quality knowledge-based plans (KBPs) using proton and photon beams to identify patients for proton treatment based on their normal tissue complication probabilities (NTCP).

Methods and Materials

Two previously validated RapidPlan PT models for locally advanced head and neck cancer were used in combination with scripting to automatically create proton and photon KBPs for 72 patients with recent oropharynx cancer. NTCPs were calculated for each patient based on the KBPs, and patient selection was simulated according to the current Dutch national protocol.

Results

The photon/proton KBP exhibited good correlation between predicted and achieved organ-at-risk mean doses, with a ≤5 Gy difference in 208/196 out of 215 structures relevant for the head and neck cancer NTCP model. The proton KBPs yielded on average 7.1/6.1/7.6 Gy lower dose to salivary/swallowing structures/oral cavity than the photon KBPs. This reduced average grade 2/3 dysphagia and xerostomia by 7.1/3.3 and 5.5/2.0 percentage points, resulting in 16 of 72 patients (22%) being indicated for proton treatment. The entire automated process took <30 minutes per patient.

Conclusions

Automated support for decision making using KBP is feasible and fast. The planning solution has potential to speed up the planning and patient-selection process significantly without major compromises to the plan quality.

Management and work-up procedures of patients with head and neck malignancies treated by radiation

Radiotherapy alone or in association with systemic treatment plays a major role in the treatment of head and neck tumours, either as a primary treatment or as a postoperative modality. The management of these tumours is multidisciplinary, requiring particular care at every treatment step. We present the update of the recommendations of the French Society of Radiation Oncology on the radiotherapy of head and neck tumours from the imaging work-up needed for optimal selection of treatment volume, to optimization of the dose distribution and delivery.

Knowledge-based dose prediction models to inform gynecologic brachytherapy needle supplementation for locally advanced cervical cancer

PURPOSE

The use of interstitial needles, combined with intracavitary applicators, enables customized dose distributions and is beneficial for complex cases, but increases procedure time. Overall, applicator selection is not standardized and depends on physician expertise and preference. The purpose of this study is to determine whether dose prediction models can guide needle supplementation decision-making for cervical cancer.

MATERIALS AND METHODS

Intracavitary knowledge-based models for organ-at-risk (OAR) dose estimation were trained and validated for tandem-and-ring/ovoids (T&R/T&O) implants. Models were applied to hybrid cases with 1-3 implanted needles to predict OAR dose without needles. As a reference, 70/67 hybrid T&R/T&O cases were replanned without needles, following a standardized procedure guided by dose predictions. If a replanned dose exceeded the dose objective, the case was categorized as requiring needles. Receiver operating characteristic (ROC) curves of needle classification accuracy were generated. Optimal classification thresholds were determined from the Youden Index.

RESULTS

Needle supplementation reduced dose to OARs. However, 67%/39% of replans for T&R/T&O met all dose constraints without needles. The ROC for T&R/T&O models had an area-under-curve of 0.89/0.86, proving high classification accuracy. The optimal threshold of 99%/101% of the dose limit for T&R/T&O resulted in classification sensitivity and specificity of 78%/86% and 85%/78%.

CONCLUSIONS

Needle supplementation reduced OAR dose for most cases but was not always required to meet standard dose objectives, particularly for T&R cases. Our knowledge-based dose prediction model accurately identified cases that could have met constraints without needle supplementation, suggesting that such models may be beneficial for applicator selection.

Evaluation of dose differences between intracavitary applicators for cervical brachytherapy using knowledge-based models

PURPOSE

Currently, there is a lack of patient-specific tools to guide brachytherapy planning and applicator choice for cervical cancer. The purpose of this study is to evaluate the accuracy of organ-at-risk (OAR) dose predictions using knowledge-based intracavitary models, and the use of these models and clinical data to determine the dosimetric differences of tandem-and-ring (T&R) and tandem-and-ovoids (T&O) applicators.

MATERIALS AND METHODS

Knowledge-based models, which predict organ D_2cc, were trained on 77/75 cases and validated on 32/38 for T&R/T&O applicators. Model performance was quantified using ΔD_2cc=D_2cc,actual-D_{2cc,predicted}, with standard deviation (σ(ΔD_2cc)) representing precision. Model-predicted applicator dose differences were determined by applying T&O models to T&R cases, and vice versa, and compared to clinically-achieved D_2cc differences. Applicator differences were assessed using a Student's t-test (p < 0.05 significant).

RESULTS

Validation T&O/T&R model precision was 0.65/0.55 Gy, 0.55/0.38 Gy, and 0.43/0.60 Gy for bladder, rectum and sigmoid, respectively, and similar to training. When applying T&O/T&R models to T&R/T&O cases, bladder, rectum and sigmoid D_2cc values in EQD2 were on average 5.69/2.62 Gy, 7.31/6.15 Gy and 3.65/0.69 Gy lower for T&R, with similar HRCTV volume and coverage. Clinical data also showed lower T&R OAR doses, with mean EQD2 D_2cc deviations of 0.61 Gy, 7.96 Gy (p < 0.01) and 5.86 Gy (p < 0.01) for bladder, rectum and sigmoid.

CONCLUSIONS

Accurate knowledge-based dose prediction models were developed for two common intracavitary applicators. These models could be beneficial for standardizing and improving the quality of brachytherapy plans. Both models and clinical data suggest that significant OAR sparing can be achieved with T&R over T&O applicators, particularly for the rectum.

Personalising treatment plan quality review with knowledge-based planning in the TROG 15.03 trial for stereotactic ablative body radiotherapy in primary kidney cancer

INTRODUCTION

Quality assurance (QA) of treatment plans in clinical trials improves protocol compliance and patient outcomes. Retrospective use of knowledge-based-planning (KBP) in clinical trials has demonstrated improved treatment plan quality and consistency. We report the results of prospective use of KBP for real-time QA of treatment plan quality in the TROG 15.03 FASTRACK II trial, which evaluates efficacy of stereotactic ablative body radiotherapy (SABR) for kidney cancer.

METHODS

A KBP model was generated based on single institution data. For each patient in the KBP phase (open to the last 31 patients in the trial), the treating centre submitted treatment plans 7 days prior to treatment. A treatment plan was created by using the KBP model, which was compared with the submitted plan for each organ-at-risk (OAR) dose constraint. A report comparing each plan for each OAR constraint was provided to the submitting centre within 24 h of receiving the plan. The centre could then modify the plan based on the KBP report, or continue with the existing plan.

RESULTS

Real-time feedback using KBP was provided in 24/31 cases. Consistent plan quality was in general achieved between KBP and the submitted plan. KBP review resulted in replan and improvement of OAR dosimetry in two patients. All centres indicated that the feedback was a useful QA check of their treatment plan.

CONCLUSION

KBP for real-time treatment plan review was feasible for 24/31 cases, and demonstrated ability to improve treatment plan quality in two cases. Challenges include integration of KBP feedback into clinical timelines, interpretation of KBP results with respect to clinical trade-offs, and determination of appropriate plan quality improvement criteria.

Automation in radiotherapy treatment planning: Examples of use in clinical practice and future trends for a complete automated workflow

Modern radiotherapy treatment planning is a complex and time-consuming process that requires the skills of experienced users to obtain quality plans. Since the early 2000s, the automation of this planning process has become an important research topic in radiotherapy. Today, the first commercial automated treatment planning solutions are available and implemented in a growing number of clinical radiotherapy departments. It should be noted that these various commercial solutions are based on very different methods, implying a daily practice that varies from one center to another. It is likely that this change in planning practices is still in its infancy. Indeed, the rise of artificial intelligence methods, based in particular on deep learning, has recently revived research interest in this subject. The numerous articles currently being published announce a lasting and profound transformation of radiotherapy planning practices in the years to come. From this perspective, an evolution of initial training for clinical teams and the drafting of new quality assurance recommendations is desirable.

Reducing variability among treatment machines using knowledge-based planning for head and neck, pancreatic, and rectal cancer

Purpose

This study aimed to assess dosimetric indices of RapidPlan model‐based plans for different energies (6, 8, 10, and 15 MV; 6‐ and 10‐MV flattening filter‐free), multileaf collimator (MLC) types (Millennium 120, High Definition 120, dual‐layer MLC), and disease sites (head and neck, pancreatic, and rectal cancer) and compare these parameters with those of clinical plans.

Methods

RapidPlan models in the Eclipse version 15.6 were used with the data of 28, 42, and 20 patients with head and neck, pancreatic, and rectal cancer, respectively. RapidPlan models of head and neck, pancreatic, and rectal cancer were created for TrueBeam STx (High Definition 120) with 6 MV, TrueBeam STx with 10‐MV flattening filter‐free, and Clinac iX (Millennium 120) with 15 MV, respectively. The models were used to create volumetric‐modulated arc therapy plans for a 10‐patient test dataset using all energy and MLC types at all disease sites. The Holm test was used to compare multiple dosimetric indices in different treatment machines and energy types.

Results

The dosimetric indices for planning target volume and organs at risk in RapidPlan model‐based plans were comparable to those in the clinical plan. Furthermore, no dose difference was observed among the RapidPlan models. The variability among RapidPlan models was consistent regardless of the treatment machines, MLC types, and energy.

Conclusions

Dosimetric indices of RapidPlan model‐based plans appear to be comparable to the ones based on clinical plans regardless of energies, MLC types, and disease sites. The results suggest that the RapidPlan model can generate treatment plans independent of the type of treatment machine.

Investigating the potential of deep learning for patient-specific quality assurance of salivary gland contours using EORTC-1219-DAHANCA-29 clinical trial data

INTRODUCTION

Manual quality assurance (QA) of radiotherapy contours for clinical trials is time and labor intensive and subject to inter-observer variability. Therefore, we investigated whether deep-learning (DL) can provide an automated solution to salivary gland contour QA.

MATERIAL AND METHODS

DL-models were trained to generate contours for parotid (PG) and submandibular glands (SMG). Sørensen-Dice coefficient (SDC) and Hausdorff distance (HD) were used to assess agreement between DL and clinical contours and thresholds were defined to highlight cases as potentially sub-optimal. 3 types of deliberate errors (expansion, contraction and displacement) were gradually applied to a test set, to confirm that SDC and HD were suitable QA metrics. DL-based QA was performed on 62 patients from the EORTC-1219-DAHANCA-29 trial. All highlighted contours were visually inspected.

RESULTS

Increasing the magnitude of all 3 types of errors resulted in progressively severe deterioration/increase in average SDC/HD. 19/124 clinical PG contours were highlighted as potentially sub-optimal, of which 5 (26%) were actually deemed clinically sub-optimal. 2/19 non-highlighted contours were false negatives (11%). 15/69 clinical SMG contours were highlighted, with 7 (47%) deemed clinically sub-optimal and 2/15 non-highlighted contours were false negatives (13%). For most incorrectly highlighted contours causes for low agreement could be identified.

CONCLUSION

Automated DL-based contour QA is feasible but some visual inspection remains essential. The substantial number of false positives were caused by sub-optimal performance of the DL-model. Improvements to the model will increase the extent of automation and reliability, facilitating the adoption of DL-based contour QA in clinical trials and routine practice.

The benefits evaluation of abdominal deep inspiration breath hold based on knowledge-based radiotherapy treatment planning for left-sided breast cancer

Purpose

To study the impact of abdominal deep inspiration breath hold (DIBH) technique on knowledge‐based radiotherapy treatment planning for left‐sided breast cancer to guide the application of DIBH technology.

Materials and methods

Two kernel density estimation (KDE) models were developed based on 40 left‐sided breast cancer patients with two CT acquisitions of free breathing (FB‐CT) and DIBH (DIBH‐CT). Each KDE model was used to predict dose volume histograms (DVHs) based on DIBH‐CT and FB‐CT for another 10 new patients similar to our training datasets. The predicted DVHs were taken as a substitute for dose constraints and objective functions in the Eclipse treatment planning system, with the same requirements for the planning target volume (PTV). The mean doses to the heart, the left anterior descending coronary artery (LADCA) and the ipsilateral lung were evaluated and compared using the T‐test among clinical plans, KDE predictions, and KDE plans.

Results

Our study demonstrated that the KDE model can generate deliverable simulations equivalent to clinically applicable plans. The T‐test was applied to test the consistency hypothesis on another ten left‐sided breast cancer patients. In cases of the same breathing status, there was no statistically significant difference between the predicted and the clinical plans for all clinically relevant DVH indices (P > 0.05), and all predicted DVHs can be transferred into deliverable plans. For DIBH‐CT images, significant differences were observed between FB model predictions and clinical plans (P < 0.05). DIBH model prediction cannot be optimized to a deliverable plan based on FB‐CT, with a counsel of perfection.

Conclusion

KDE models can predict DVHs well for the same breathing conditions but degrade with different breathing conditions. The benefits of DIBH for a given patient can be evaluated with a quick comparison of prediction results of the two models before treatment planning.

Offline Quality Assurance for Intensity Modulated Radiation Therapy Treatment Plans for NRG-HN001 Head and Neck Clinical Trial Using Knowledge-Based Planning

Purpose

This study aimed to investigate whether a disease site–specific, multi-institutional knowledge based-planning (KBP) model can improve the quality of intensity modulated radiation therapy treatment planning for patients enrolled in the head and neck NRG-HN001clinical trial and to establish a threshold of improvements of treatment plans submitted to the clinical trial.

Methods and Materials

Fifty treatment plans for patients enrolled in the NRG-HN001 clinical trial were used to build a KBP model; the model was then used to reoptimize 50 other plans. We compared the dosimetric parameters of the submitted and KBP reoptimized plans. We compared differences between KBP and submitted plans for single- and multi-institutional treatment plans.

Results

Mean values for the dose received by 95% of the planning target volume (PTV_6996) and for the maximum dose (D0.03cc) of PTV_6996 were 0.5 Gy and 2.1 Gy higher in KBP plans than in the submitted plans, respectively. Mean values for D0.03cc to the brain stem, spinal cord, optic nerve_R, optic nerve_L, and chiasm were 2.5 Gy, 1.9 Gy, 6.4 Gy, 6.6 Gy, and 5.7 Gy lower in the KBP plans than in the submitted plans. Mean values for D_mean to parotid_R and parotid_L glands were 2.2 Gy and 3.8 Gy lower in KBP plans, respectively. In 33 out of 50 KBP plans, we observed improvements in sparing of at least 7 organs at risk (OARs) (brain stem, spinal cord, optic nerves (R & L), chiasm, and parotid glands [R & L]). A threshold of improvement of OARs sparing of 5% of the prescription dose was established for providing the quality assurance results back to the treating institution.

Conclusions

A disease site–specific, multi-institutional, clinical trial-based KBP model improved sparing of OARs in a large number of reoptimized plans submitted to the NRG-HN001 clinical trial, and the model is being used as an offline quality assurance tool.

Robust dose planning objectives for mesorectal radiotherapy of early stage rectal cancer - A multicentre dose planning study

BACKGROUND AND PURPOSE

Organ preservation strategies are increasingly being explored for early rectal cancer. This requires revision of target volumes according to disease stage, as well as new guidelines for treatment planning. We conducted an international, multicentre dose planning study to develop robust planning objectives for modern radiotherapy of a novel mesorectal-only target volume, as implemented in the STAR-TReC trial (NCT02945566).

MATERIALS AND METHODS

The published literature was used to establish relevant dose levels for organ at risk (OAR) plan optimisation. Ten representative patients with early rectal cancer were identified. Treatment scans had mesorectal target volumes as well as bowel cavity, bladder and femoral heads outlined, and were circulated amongst the three participating institutions. Each institution produced plans for short course (SCRT, 5 × 5 Gy) and long course (LCRT, 25 × 2 Gy) treatment, using volumetric modulated arc therapy on different dose planning systems. Optimisation objectives for OARs were established by determining dose metric objectives achievable for ≥90% of plans.

RESULTS

Sixty plans, all fulfilling target coverage criteria, were produced. The planning results and literature review suggested optimisation objectives for SCRT: V 10Gy < 180 cm3, V 18Gy < 110 cm3, V 23Gy < 85 cm3 for bowel cavity; V 21Gy < 15% and V 25Gy < 5% for bladder; and V 12.5Gy < 11% for femoral heads. Corresponding objectives for LCRT: V 20Gy < 180 cm3, V 30Gy < 130 cm3, V 45Gy < 90 cm3 for bowel cavity; V 35Gy < 22% and V 50Gy < 7% for bladder; and V 25Gy < 15% for femoral heads. Constraints were validated across all three institutions.

CONCLUSION

We utilized a multicentre planning study approach to develop robust planning objectives for mesorectal radiotherapy for early rectal cancer.

Deep Learning-Based Delineation of Head and Neck Organs at Risk: Geometric and Dosimetric Evaluation

PURPOSE

Organ-at-risk (OAR) delineation is a key step in treatment planning but can be time consuming, resource intensive, subject to variability, and dependent on anatomical knowledge. We studied deep learning (DL) for automated delineation of multiple OARs; in addition to geometric evaluation, the dosimetric impact of using DL contours for treatment planning was investigated.

METHODS AND MATERIALS

The following OARs were delineated with DL developed in-house: both submandibular and parotid glands, larynx, cricopharynx, pharyngeal constrictor muscle (PCM), upper esophageal sphincter, brain stem, oral cavity, and esophagus. DL contours were benchmarked against the manual delineation (MD) clinical contours using the Sørensen-Dice similarity coefficient. Automated knowledge-based treatment plans were used. The mean dose to the manually delineated OAR structures was reported for the MD and DL plans.

RESULTS

DL delineation of all OARs took <10 seconds per patient. For 7 of 11 OARs, the average Sørensen-Dice similarity coefficient was good (0.78-0.83). However, performance was lower for the esophagus (0.60), brainstem (0.64), PCM (0.68), and cricopharynx (0.73), often because of variations in MD. Although the average dose was statistically significantly higher in the DL plans for the inferior PCM (1.4 Gy) and esophagus (2.2 Gy), these average differences were not clinically significant. Dose to 28 of 209 (13.4%) and 7 of 209 (3.3%) OARs was >2 Gy higher and >2 Gy lower, respectively, in the DL plans.

CONCLUSIONS

DL-based segmentation for head and neck OARs is fast; for most organs and most patients, it performs sufficiently well for treatment-planning purposes. It has the potential to increase efficiency and facilitate online adaptive radiation therapy.