Multi-criteria optimization achieves superior normal tissue sparing in intensity-modulated radiation therapy for oropharyngeal cancer patients

Multicriteria optimization of radiation therapy: Towards empowerment and standardization of reverse planning for head and neck squamous cell carcinoma

PURPOSE

The purpose of this study was to assess if multicriteria optimization could limit interoperator variability in radiation therapy planning and assess if this method could contribute to target volume coverage and sparing of organ at risk for intensity-modulated curative radiation therapy of head and neck cancers.

MATERIAL AND METHODS

We performed a retrospective analysis on 20 patients treated for an oropharyngeal or oral cavity squamous cell carcinoma. We carried out a comparative dosimetric study of manual plans produced with Precision® software, compared with the plans proposed using the multicriteria optimization method (RayStation®). We assessed interoperator reproducibility on the first six patients, and dosimetric contribution in sparing organs at risk using the multicriteria optimization method.

RESULTS

Median age was 69 years, most lesions were oropharyngeal carcinoma (65%), and 35% lesions were stage T3. First, we obtained a high degree of similarity between the four operator measurements for each patient at the level of each organ. Intraclass correlation coefficients were greater than 0.85. Second, we observed a significant dosimetric benefit for contralateral parotid gland, homolateral and contralateral masseter muscles, homolateral and contralateral pterygoid muscles and for the larynx (P<0.05). For the contralateral parotid gland, the mean dose difference between the multicriteria optimization and manual plans was -2.0Gy (P=0.01). Regarding the larynx, the mean dose difference between the two plans was -4.6Gy (P<0.001).

CONCLUSION

Multicriteria optimization is a reproducible technique and faster than manual optimization. It allows dosimetric advantages on organs at risk, especially for those not usually taken into consideration in manual dosimetry. This may lead to improved quality of life.

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.

Comparative Study of Eclipse and RayStation Multi-Criteria Optimization-Based Prostate Radiotherapy Treatment Planning Quality

Multi-criteria optimization (MCO) function has been available on commercial radiotherapy (RT) treatment planning systems to improve plan quality; however, no study has compared Eclipse and RayStation MCO functions for prostate RT planning. The purpose of this study was to compare prostate RT MCO plan qualities in terms of discrepancies between Pareto optimal and final deliverable plans, and dosimetric impact of final deliverable plans. In total, 25 computed tomography datasets of prostate cancer patients were used for Eclipse (version 16.1) and RayStation (version 12A) MCO-based plannings with doses received by 98% of planning target volume having 76 Gy prescription (PTV76D98%) and 50% of rectum (rectum D50%) selected as trade-off criteria. Pareto optimal and final deliverable plan discrepancies were determined based on PTV76D98% and rectum D50% percentage differences. Their final deliverable plans were compared in terms of doses received by PTV76 and other structures including rectum, and PTV76 homogeneity index (HI) and conformity index (CI), using a t-test. Both systems showed discrepancies between Pareto optimal and final deliverable plans (Eclipse: -0.89% (PTV76D98%) and -2.49% (Rectum D50%); RayStation: 3.56% (PTV76D98%) and -1.96% (Rectum D50%)). Statistically significantly different average values of PTV76D98%,HI and CI, and mean dose received by rectum (Eclipse: 76.07 Gy, 0.06, 1.05 and 39.36 Gy; RayStation: 70.43 Gy, 0.11, 0.87 and 51.65 Gy) are noted, respectively (p < 0.001). Eclipse MCO-based prostate RT plan quality appears better than that of RayStation.

Clinical benefit of range uncertainty reduction in proton treatment planning based on dual-energy CT for neuro-oncological patients

OBJECTIVE

Several studies have shown that dual-energy CT (DECT) can lead to improved accuracy for proton range estimation. This study investigated the clinical benefit of reduced range uncertainty, enabled by DECT, in robust optimisation for neuro-oncological patients.

METHODS

DECT scans for 27 neuro-oncological patients were included. Commercial software was applied to create stopping-power ratio (SPR) maps based on the DECT scan. Two plans were robustly optimised on the SPR map, keeping the beam and plan settings identical to the clinical plan. One plan was robustly optimised and evaluated with a range uncertainty of 3% (as used clinically; denoted 3%-plan); the second plan applied a range uncertainty of 2% (2%-plan). Both plans were clinical acceptable and optimal. The dose-volume histogram parameters were compared between the two plans. Two experienced neuro-radiation oncologists determined the relevant dose difference for each organ-at-risk (OAR). Moreover, the OAR toxicity levels were assessed.

RESULTS

For 24 patients, a dose reduction >0.5/1 Gy (relevant dose difference depending on the OAR) was seen in one or more OARs for the 2%-plan; e.g. for brainstem D0.03cc in 10 patients, and hippocampus D40% in 6 patients. Furthermore, 12 patients had a reduction in toxicity level for one or two OARs, showing a clear benefit for the patient.

CONCLUSION

Robust optimisation with reduced range uncertainty allows for reduction of OAR toxicity, providing a rationale for clinical implementation. Based on these results, we have clinically introduced DECT-based proton treatment planning for neuro-oncological patients, accompanied with a reduced range uncertainty of 2%.

ADVANCES IN KNOWLEDGE

This study shows the clinical benefit of range uncertainty reduction from 3% to 2% in robustly optimised proton plans. A dose reduction to one or more OARs was seen for 89% of the patients, and 44% of the patients had an expected toxicity level decrease.

Input feature design and its impact on the performance of deep learning models for predicting fluence maps in intensity-modulated radiation therapy

Objective. Deep learning (DL) models for fluence map prediction (FMP) have great potential to reduce treatment planning time in intensity-modulated radiation therapy (IMRT) by avoiding the lengthy inverse optimization process. This study aims to improve the rigor of input feature design in a DL-FMP model by examining how different designs of input features influence model prediction performance.Approach. This study included 231 head-and-neck intensity-modulated radiation therapy patients. Three input feature designs were investigated. The first design (D1) assumed that information of all critical structures from all beam angles should be combined to predict fluence maps. The second design (D2) assumed that local anatomical information was sufficient for predicting radiation intensity of a beamlet at a respective beam angle. The third design (D3) assumed the need for both local anatomical information and inter-beam modulation to predict radiation intensity values of the beamlets that intersect at a voxel. For each input design, we tailored the DL model accordingly. All models were trained using the same set of ground truth plans (GT plans). The plans generated by DL models (DL plans) were analyzed using key dose-volume metrics. One-way ANOVA with multiple comparisons correction (Bonferroni method) was performed (significance level = 0.05).Main results. For PTV-related metrics, all DL plans had significantly higher maximum dose (p < 0.001), conformity index (p < 0.001), and heterogeneity index (p < 0.001) compared to GT plans, with D2 being the worst performer. Meanwhile, except for cord+5 mm (p < 0.001), DL plans of all designs resulted in OAR dose metrics that are comparable to those of GT plans.Significance. Local anatomical information contains most of the information that DL models need to predict fluence maps for clinically acceptable OAR sparing. Input features from beam angles are needed to achieve the best PTV coverage. These results provide valuable insights for further improvement of DL-FMP models and DL models in general.

Data-Driven Dose-Volume Histogram Prediction

Purpose

To evaluate dose-volume histogram (DVH) prediction from prior radiation therapy data.

Methods and Materials

An Oncospace radiation therapy database was constructed including images, structures, and dose distributions for patients with advanced lung cancer. DVH data was queried for total lungs, esophagus, heart, and external body contours. Each query returned DVH data for the N-most similar organs at risk (OARs) based on OAR-to-planning-target-volume (PTV) geometry via the overlap volume histogram (OVH). The DVHs for 5, 20, and 50 of the most similar OVHs were returned for each OAR for each patient. The OVH(0cm) is the relative volume of the OAR overlapping with the PTV, and the OVH(2cm) is the relative volume of the OAR 2 cm away from the PTV. The OVH(cm) and DVH(%) queried from the database were separated into interquartile ranges (IQRs), nonoutlier ranges (NORs) (equal to 3 × IQR), and the average database DVH (DVH-DB) computed from the NOR data. The ability to predict the clinically delivered DVH was evaluated based on percentiles and differences between the DVH-DB and the clinical DVH (DVH-CL) for a varying number of returned patient DVHs for a subset of patients.

Results

The ability to predict the clinically delivered DVH was excellent in the lungs and body; the IQR and NOR were <4% and <16%, respectively, in the lungs and <1% and <5%, respectively, in the body at all distances less than 2 cm from the PTV. For 21/23 patients considered, the differences in lung DVH-DB and DVH-CL were <4.6% and in 14/23 cases, <3%. In esophagus and heart, the ability to predict DVH-CL was weaker, with mean DVH differences >10% for 12/23 esophagi and 10/23 hearts. In esophagus and heart queries, the NOR was often 10% to 100% volume in dose ranges between 0% and 50% of prescription, independent of the number of patients queried.

Conclusions

Using prior data to predict clinical dosimetry is increasingly of interest, but model- and data-driven methods have limitations if based on limited data sets. This study's results showed that prediction may be reasonable in organs containing tumors with known overlap, but for nonoverlapped OARs, planning preference and plan design may dominate the clinical dose.

Insights of an AI agent via analysis of prediction errors: a case study of fluence map prediction for radiation therapy planning

Purpose.We have previously reported an artificial intelligence (AI) agent that automatically generates intensity-modulated radiation therapy (IMRT) plans via fluence map prediction, by-passing inverse planning. This AI agent achieved clinically comparable quality for prostate cases, but its performance on head-and-neck patients leaves room for improvement. This study aims to collect insights of the deep-learning-based (DL-based) fluence map prediction model by systematically analyzing its prediction errors.Methods.From the modeling perspective, the DL model's output is the fluence maps of IMRT plans. However, from the clinical planning perspective, the plan quality evaluation should be based on the clinical dosimetric criteria such as dose-volume histograms. To account for the complex and non-intuitive relationships between fluence map prediction errors and the corresponding dose distribution changes, we propose a novel error analysis approach that systematically examines plan dosimetric changes that are induced by varying amounts of fluence prediction errors. We investigated four decomposition modes of model prediction errors. The two spatial domain decompositions are based on fluence intensity and fluence gradient. The two frequency domain decompositions are based on Fourier-space banded frequency rings and Fourier-space truncated low-frequency disks. The decomposed error was analyzed for its impact on the resulting plans' dosimetric metrics. The analysis was conducted on 15 test cases spared from the 200 training and 16 validation cases used to train the model.Results.Most planning target volume metrics were significantly correlated with most error decompositions. The Fourier space disk radii had the largest Spearman's coefficients. The low-frequency region within a disk of ∼20% Fourier space contained most of errors that impact overall plan quality.Conclusions.This study demonstrates the feasibility of using fluence map prediction error analysis to understand the AI agent's performance. Such insights will help fine-tune the DL models in architecture design and loss function selection.

Influence of target dose heterogeneity on dose sparing of normal tissue in peripheral lung tumor stereotactic body radiation therapy

OBJECTIVE

To evaluate the influence of target dose heterogeneity on normal tissue dose sparing for peripheral lung tumor stereotactic body radiation therapy (SBRT).

METHODS

Based on the volumetric-modulated arc therapy (VMAT) technique, three SBRT plans with homogeneous, moderate heterogeneous, and heterogeneous (HO, MHE, and HE) target doses were compared in 30 peripheral lung tumor patients. The prescription dose was 48 Gy in 4 fractions. Ten rings outside the PTV were created to limit normal tissue dosage and evaluate dose falloff.

RESULTS

When MHE and HE plans were compared to HO plans, the conformity index of the PTV was increased by approximately 0.08. The median mean lung dose (MLD), V₅, V₁₀, V₂₀ of whole lung, D_2%, D_1cc, D_2cc of the rib, V₃₀ of the rib, D_2% and the maximum dose (D_max) of the skin, and D_2% and D_max of most mediastinal organs at risk (OARs) and spinal cord were reduced by up to 4.51 Gy or 2.8%. Analogously, the median D_max, D_2% and mean dose of rings were reduced by 0.71 to 8.46 Gy; and the median R_50% and D_2cm were reduced by 2.1 to 2.3 and 7.4% to 8.0%, respectively. Between MHE and HE plans there was little to no difference in OARs dose and dose falloff beyond the target. Furthermore, the dose sparing of rib V₃₀ and the mean dose of rings were negatively correlated with the rib and rings distance from tumor, respectively.

CONCLUSIONS

For peripheral lung tumor SBRT, target conformity, normal tissue dose, and dose falloff around the target could be improved by loosening or abandoning homogeneity. While there was negligible further dose benefit for the maximum target dose above 125% of the prescription, dose sparing of normal tissue derived from a heterogeneous target decreased as the distance from the tumor increased.

Dosimetric Quality of Online Adapted Pancreatic Cancer Treatment Plans on an MRI-Guided Radiation Therapy System

Purpose

Stereotactic magnetic resonance image–guided adaptive radiation therapy (SMART) is an emerging technique that shows promise in the treatment of pancreatic cancer and other abdominopelvic malignancies. However, it is unknown whether the time-limited nature of on-table adaptive planning may result in dosimetrically suboptimal plans. The purpose of this study was to quantitatively address that question through systemic retrospective replanning of treated on-table adaptive pancreatic cancer cases.

Methods and Materials

Of 74 consecutive adapted fractions, 30 were retrospectively replanned based on deficiencies in planning target volume (PTV) and gross tumor volume (GTV) coverage or doses to organs-at-risk (OARs) that exceeded ideal constraints. Retrospective plans were created by adjusting dose-volume objectives in an iterative fashion until deemed optimized. The goal of replanning was to improve PTV/GTV coverage while keeping the dose to gastrointestinal OARs the same or lower or to reduce OAR doses while keeping PTV coverage the same or higher. The global maximum dose was required to be maintained within 2% of that of the treated adaptive plan to eliminate it as a confounding factor. A threshold of 5% improvement in PTV coverage or 5% decrease in OAR dose was used to define a clinically significant improvement.

Results

Of the 30 replans, 7 obtained at least 5% PTV coverage improvement. The average increase in PTV coverage for these plans was 11%. No plans were clinically significantly improved in terms of OAR sparing. Changes in beam-on time did not show any correlation. Statistical analysis via a linear mixed-effects model with a nested random effect suggested that both GTV and PTV coverage were improved over SMART process plans by 0.91 cc (P = .02) and 2.03 cc (P < .001), respectively.

Conclusions

Dosimetric plan quality of at least 10% of SMART fractions may be improved through more extensive replanning than is currently performed on-table. Further work is needed to develop an automated replanning workflow to streamline the in-depth replanning process to better fit into an on-table adaptive workflow.

An artificial intelligence-driven agent for real-time head-and-neck IMRT plan generation using conditional generative adversarial network (cGAN)

PURPOSE

To develop an artificial intelligence (AI) agent for fully automated rapid head-and-neck intensity-modulated radiation therapy (IMRT) plan generation without time-consuming dose-volume-based inverse planning.

METHODS

This AI agent was trained via implementing a conditional generative adversarial network (cGAN) architecture. The generator, PyraNet, is a novel deep learning network that implements 28 classic ResNet blocks in pyramid-like concatenations. The discriminator is a customized four-layer DenseNet. The AI agent first generates multiple customized two-dimensional projections at nine template beam angles from a patient's three-dimensional computed tomography (CT) volume and structures. These projections are then stacked as four-dimensional inputs of PyraNet, from which nine radiation fluence maps of the corresponding template beam angles are generated simultaneously. Finally, the predicted fluence maps are automatically postprocessed by Gaussian deconvolution operations and imported into a commercial treatment planning system (TPS) for plan integrity check and visualization. The AI agent was built and tested upon 231 oropharyngeal IMRT plans from a TPS plan library. 200/16/15 plans were assigned for training/validation/testing, respectively. Only the primary plans in the sequential boost regime were studied. All plans were normalized to 44 Gy prescription (2 Gy/fx). A customized Harr wavelet loss was adopted for fluence map comparison during the training of the PyraNet. For test cases, isodose distributions in AI plans and TPS plans were qualitatively evaluated for overall dose distributions. Key dosimetric metrics were compared by Wilcoxon signed-rank tests with a significance level of 0.05.

RESULTS

All 15 AI plans were successfully generated. Isodose gradients outside of PTV in AI plans were comparable to those of the TPS plans. After PTV coverage normalization, D_mean of left parotid (D_AI  = 23.1 ± 2.4 Gy; D_TPS  = 23.1 ± 2.0 Gy), right parotid (D_AI  = 23.8 ± 3.0 Gy; D_TPS  = 23.9 ± 2.3 Gy), and oral cavity (D_AI  = 24.7 ± 6.0 Gy; D_TPS  = 23.9 ± 4.3 Gy) in the AI plans and the TPS plans were comparable without statistical significance. AI plans achieved comparable results for maximum dose at 0.01cc of brainstem (D_AI  = 15.0 ± 2.1 Gy; D_TPS  = 15.5 ± 2.7 Gy) and cord + 5mm (D_AI  = 27.5 ± 2.3 Gy; D_TPS  = 25.8 ± 1.9 Gy) without clinically relevant differences, but body D_max results (D_AI  = 121.1 ± 3.9 Gy; D_TPS  = 109.0 ± 0.9 Gy) were higher than the TPS plan results. The AI agent needed ~3 s for predicting fluence maps of an IMRT plan.

CONCLUSIONS

With rapid and fully automated execution, the developed AI agent can generate complex head-and-neck IMRT plans with acceptable dosimetry quality. This approach holds great potential for clinical applications in preplanning decision-making and real-time planning.