skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Evaluation of a Knowledge-Based Planning Solution for Head and Neck Cancer

Abstract

Purpose: Automated and knowledge-based planning techniques aim to reduce variations in plan quality. RapidPlan uses a library consisting of different patient plans to make a model that can predict achievable dose-volume histograms (DVHs) for new patients and uses those models for setting optimization objectives. We benchmarked RapidPlan versus clinical plans for 2 patient groups, using 3 different libraries. Methods and Materials: Volumetric modulated arc therapy plans of 60 recent head and neck cancer patients that included sparing of the salivary glands, swallowing muscles, and oral cavity were evenly divided between 2 models, Model{sub 30A} and Model{sub 30B}, and were combined in a third model, Model{sub 60}. Knowledge-based plans were created for 2 evaluation groups: evaluation group 1 (EG1), consisting of 15 recent patients, and evaluation group 2 (EG2), consisting of 15 older patients in whom only the salivary glands were spared. RapidPlan results were compared with clinical plans (CP) for boost and/or elective planning target volume homogeneity index, using HI{sub B}/HI{sub E} = 100 × (D2% − D98%)/D50%, and mean dose to composite salivary glands, swallowing muscles, and oral cavity (D{sub sal}, D{sub swal}, and D{sub oc}, respectively). Results: For EG1, RapidPlan improved HI{sub B} and HI{sub E} values compared with CP by 1.0% tomore » 1.3% and 1.0% to 0.6%, respectively. Comparable D{sub sal} and D{sub swal} values were seen in Model{sub 30A}, Model{sub 30B}, and Model{sub 60}, decreasing by an average of 0.1, 1.0, and 0.8 Gy and 4.8, 3.7, and 4.4 Gy, respectively. However, differences were noted between individual organs at risk (OARs), with Model{sub 30B} increasing D{sub oc} by 0.1, 3.2, and 2.8 Gy compared with CP, Model{sub 30A}, and Model{sub 60}. Plan quality was less consistent when the patient was flagged as an outlier. For EG2, RapidPlan decreased D{sub sal} by 4.1 to 4.9 Gy on average, whereas HI{sub B} and HI{sub E} decreased by 1.1% to 1.5% and 2.3% to 1.9%, respectively. Conclusions: RapidPlan knowledge-based treatment plans were comparable to CP if the patient's OAR-planning target volume geometry was within the range of those included in the models. EG2 results showed that a model including swallowing-muscle and oral-cavity sparing can be applied to patients with only salivary gland sparing. This may allow model library sharing between institutes. Optimal detection of inadequate plans and population of model libraries requires further investigation.« less

Authors:
; ; ; ;
Publication Date:
OSTI Identifier:
22458632
Resource Type:
Journal Article
Resource Relation:
Journal Name: International Journal of Radiation Oncology, Biology and Physics; Journal Volume: 91; Journal Issue: 3; Other Information: Copyright (c) 2015 Elsevier Science B.V., Amsterdam, The Netherlands, All rights reserved.; Country of input: International Atomic Energy Agency (IAEA)
Country of Publication:
United States
Language:
English
Subject:
62 RADIOLOGY AND NUCLEAR MEDICINE; BENCHMARKS; COMPARATIVE EVALUATIONS; DATA; GEOMETRY; HAZARDS; HEAD; LIBRARIES; MATHEMATICAL SOLUTIONS; MUSCLES; NECK; NEOPLASMS; OPTIMIZATION; ORAL CAVITY; PATIENTS; PLANNING; RADIATION DOSES; RADIOTHERAPY; SALIVARY GLANDS

Citation Formats

Tol, Jim P., E-mail: j.tol@vumc.nl, Delaney, Alexander R., Dahele, Max, Slotman, Ben J., and Verbakel, Wilko F.A.R. Evaluation of a Knowledge-Based Planning Solution for Head and Neck Cancer. United States: N. p., 2015. Web. doi:10.1016/J.IJROBP.2014.11.014.
Tol, Jim P., E-mail: j.tol@vumc.nl, Delaney, Alexander R., Dahele, Max, Slotman, Ben J., & Verbakel, Wilko F.A.R. Evaluation of a Knowledge-Based Planning Solution for Head and Neck Cancer. United States. doi:10.1016/J.IJROBP.2014.11.014.
Tol, Jim P., E-mail: j.tol@vumc.nl, Delaney, Alexander R., Dahele, Max, Slotman, Ben J., and Verbakel, Wilko F.A.R. Sun . "Evaluation of a Knowledge-Based Planning Solution for Head and Neck Cancer". United States. doi:10.1016/J.IJROBP.2014.11.014.
@article{osti_22458632,
title = {Evaluation of a Knowledge-Based Planning Solution for Head and Neck Cancer},
author = {Tol, Jim P., E-mail: j.tol@vumc.nl and Delaney, Alexander R. and Dahele, Max and Slotman, Ben J. and Verbakel, Wilko F.A.R.},
abstractNote = {Purpose: Automated and knowledge-based planning techniques aim to reduce variations in plan quality. RapidPlan uses a library consisting of different patient plans to make a model that can predict achievable dose-volume histograms (DVHs) for new patients and uses those models for setting optimization objectives. We benchmarked RapidPlan versus clinical plans for 2 patient groups, using 3 different libraries. Methods and Materials: Volumetric modulated arc therapy plans of 60 recent head and neck cancer patients that included sparing of the salivary glands, swallowing muscles, and oral cavity were evenly divided between 2 models, Model{sub 30A} and Model{sub 30B}, and were combined in a third model, Model{sub 60}. Knowledge-based plans were created for 2 evaluation groups: evaluation group 1 (EG1), consisting of 15 recent patients, and evaluation group 2 (EG2), consisting of 15 older patients in whom only the salivary glands were spared. RapidPlan results were compared with clinical plans (CP) for boost and/or elective planning target volume homogeneity index, using HI{sub B}/HI{sub E} = 100 × (D2% − D98%)/D50%, and mean dose to composite salivary glands, swallowing muscles, and oral cavity (D{sub sal}, D{sub swal}, and D{sub oc}, respectively). Results: For EG1, RapidPlan improved HI{sub B} and HI{sub E} values compared with CP by 1.0% to 1.3% and 1.0% to 0.6%, respectively. Comparable D{sub sal} and D{sub swal} values were seen in Model{sub 30A}, Model{sub 30B}, and Model{sub 60}, decreasing by an average of 0.1, 1.0, and 0.8 Gy and 4.8, 3.7, and 4.4 Gy, respectively. However, differences were noted between individual organs at risk (OARs), with Model{sub 30B} increasing D{sub oc} by 0.1, 3.2, and 2.8 Gy compared with CP, Model{sub 30A}, and Model{sub 60}. Plan quality was less consistent when the patient was flagged as an outlier. For EG2, RapidPlan decreased D{sub sal} by 4.1 to 4.9 Gy on average, whereas HI{sub B} and HI{sub E} decreased by 1.1% to 1.5% and 2.3% to 1.9%, respectively. Conclusions: RapidPlan knowledge-based treatment plans were comparable to CP if the patient's OAR-planning target volume geometry was within the range of those included in the models. EG2 results showed that a model including swallowing-muscle and oral-cavity sparing can be applied to patients with only salivary gland sparing. This may allow model library sharing between institutes. Optimal detection of inadequate plans and population of model libraries requires further investigation.},
doi = {10.1016/J.IJROBP.2014.11.014},
journal = {International Journal of Radiation Oncology, Biology and Physics},
number = 3,
volume = 91,
place = {United States},
year = {Sun Mar 01 00:00:00 EST 2015},
month = {Sun Mar 01 00:00:00 EST 2015}
}
  • Purpose: Knowledge Based Radiation Therapy Treatment (KBRT) planning can be used to semi-automatically generate IMRT plans for new patients using constraints derived from previously manually-planned, geometrically similar patients. We investigate whether KBRT plans can achieve greater dose sparing than manual plans using optimized, organspecific constraint weighting factors. Methods: KBRT planning of HNC radiotherapy cases geometrically matched each new (query) case to one of the 105 clinically approved plans in our database. The dose distribution of the planned match was morphed to fit the querys geometry. Dose-volume constraints extracted from the morphed dose distribution were used to run the IMRT optimizationmore » with no user input. In the first version, all constraints were multiplied by a weighting factor of 0.7. The weighting factors were then systematically optimized (in order of OARs with increasing separation from the target) to maximize sparing to each OAR without compromising other OARs. The optimized, second version plans were compared against the first version plans and the clinically approved plans for 45 unilateral/bilateral target cases using the dose metrics: mean, median and maximum (brainstem and cord) doses. Results: Compared to the first version, the second version significantly reduced mean/median contralateral parotid doses (>2Gy) for bilateral cases. Other changes between the two versions were not clinically meaningful. Compared to the original clinical plans, both bilateral and unilateral plans in the second version had lower average dose metrics for 5 of the 6 OARs. Compared to the original plans, the second version achieved dose sparing that was at least as good for all OARs and better for the ipsilateral parotid (bilateral) and oral cavity (bilateral/unilateral). Differences in planning target volume coverage metrics were not clinically significant. Conclusion: HNC-KBRT planning generated IMRT plans with at least equivalent dose sparing to manually generated plans; greater dose sparing was achieved in selected OARs.« less
  • Purpose: Knowledge-based Planning (KBP) founded on prior planning experience and Auto-Planning Engine (APE; commercialized in Pinnacle v9.10 TPS) based on progressive optimization algorithm both aim to eliminate the trial-and-error process in radiotherapy inverse planning. This study investigates the performance of the approaches in a multi-institutional setting to evaluate their functionalities in oropharyngeal cancer and offers suggestions how they can be implemented in the clinic. Methods: Radboud University Medical Center (RUMC) provided 35 oropharyngeal cancer patients (SIB-IMRT with two-dose-level prescription: 68 Gy to PTV68 and 50.3 Gy to PTV50.3) with corresponding comparative APE plans. Johns Hopkins University (JHU) contributed to amore » three-dose-level (70 Gy 63 Gy and 58.1 Gy) plan library for RUMC’s patient KBP generation. MedStar Georgetown University Hospital (MGUH) contributed to a KBP approach employing overlap-volume histogram (OVH-KBP) for generating RUMC’s patient KBP plans using JHU’s plan library. Since both approaches need their own user-defined parameters as initial inputs the first 10 patients were set aside as training set to finalize them. Meanwhile cross-institutional comparisons and adjustments were implemented for investigating institutions’ protocol discrepancies and the approaches’ user-defined parameters were updated accordingly. The finalized parameters were then applied to the remaining 25 patients for OVH-KBP and APE generation. A Wilcoxon rank-sum test was used for statistical comparison with significance level of p<0.05. Results: On average PTV68’s V95 was 96.5% in APE plans vs. 97% in OVH-KBP plans (p=0.36); PTV50.3’s V95 in APE plans was 97.8% vs.97.6% in OVH-KBP plans (p=0.6); cord’s D0.1 cc was 38.6 Gy in OVH-KBP plans vs. 43.7 Gy in APE plans (p=0.0001); mean doses to larynxes oral cavities parotids and submandibular glands were similar with p>0.2. Conclusions: The study demonstrates that KBP and APE can generate plans of comparable quality in a multi-institutional setting. Variations in clinical protocols can be effectively addressed for cross-institutional adaptations. Binbin Wu and Todd McNutt are the co-inventors of a patent associated with the proposed knowledge-based planning system which was licensed to Varian Medical Systems in 2015; This research was in part supported by Philips Radiation Oncology Systems.« less
  • Purpose: HNC IMRT treatment planning is a challenging process that relies heavily on the planner’s experience. Previously, we used the single, best match from a library of manually planned cases to semi-automatically generate IMRT plans for a new patient. The current multi-case Knowledge Based Radiation Therapy (MC-KBRT) study utilized different matching cases for each of six individual organs-at-risk (OARs), then combined those six cases to create the new treatment plan. Methods: From a database of 103 patient plans created by experienced planners, MC-KBRT plans were created for 40 (17 unilateral and 23 bilateral) HNC “query” patients. For each case, 2Dmore » beam’s-eye-view images were used to find similar geometric “match” patients separately for each of 6 OARs. Dose distributions for each OAR from the 6 matching cases were combined and then warped to suit the query case’s geometry. The dose-volume constraints were used to create the new query treatment plan without the need for human decision-making throughout the IMRT optimization. The optimized MC-KBRT plans were compared against the clinically approved plans and Version 1 (original KBRT) using the dose metrics: mean, median, and maximum (brainstem and cord+5mm) doses. Results: Compared to Version 1, MC-KBRT had no significant reduction of the dose to any of the OARs in either unilateral/bilateral cases. Compared to the manually-planned unilateral cases, there was significant reduction of the oral cavity mean/median dose (>2Gy) at the expense of the contralateral parotid. Compared to the manually-planned bilateral cases, reduction of dose was significant in the ipsilateral parotid, larynx, and oral cavity (>3Gy mean/median) while maintaining PTV coverage. Conclusion: MC-KBRT planning in head and neck cancer generates IMRT plans with equivalent dose sparing to manually created plans. MC-KBRT using multiple case matches does not show significant dose reduction compared to using a single match case with dose warping.« less
  • Purpose: To define the best threshold for tumor volume delineation of the (18) fluoro-2-deoxy-glucose positron emission tomography ({sup 18}FDG-PET) signal for radiotherapy treatment planning of intensity-modulated radiotherapy (IMRT) in head and neck cancer. Methods and Materials: In 25 patients with head-and-neck cancer, CT-based gross tumor volume (GTV{sub CT}) was delineated. After PET-CT image fusion, window level (L) was adapted to best fit the GTV{sub CT}, and GTV{sub PET} was delineated. Tumor maximum (S) and background uptake (B) were measured, and the threshold of the background-subtracted tumor maximum uptake (THR) was used for PET signal segmentation. Gross tumor volumes were expandedmore » to planning target volumes (PTVs) and analyzed. Results: The mean value of S was 40 kBq/mL, S/B ratio was 16, and THR was 26%. The THR correlated with S (r = -0.752), but no correlation between THR and the S/B ratio was seen (r = -0.382). In 77% of cases, S was >30 kBq/mL, and in 23% it was {<=}30 kBq/mL, with a mean THR of 21.4% and 41.6%, respectively (p < 0.001). Using PTV{sub PET} in radiotherapy treatment planning resulted in a reduced PTV in 72% of cases, while covering 88.2% of GTV{sub CT}, comparable to the percentage of GTV{sub PET} covered by PTV{sub CT} (p = 0.15). Conclusions: A case-specific PET signal threshold is optimal in PET-based radiotherapy treatment planning. Signal gating using a THR of 20% in tumors with S >30% {+-} 1.6% kBq/mL and 40% in tumors with S {<=}30% {+-} 1.6% kBq/mL is suitable.« less
  • Purpose: Adaptive Radiotherapy (ART) with frequent CT imaging has been used to improve dosimetric accuracy by accounting for anatomical variations, such as primary tumor shrinkage and/or body weight loss, in Head and Neck (H&N) patients. In most ART strategies, the difference between the planned and the delivered dose is estimated by generating new plans on repeated CT scans using dose-volume constraints used with the initial planning CT without considering already delivered dose. The aim of this study was to assess the dosimetric gains achieved by re-planning based on prior dose by comparing them to re-planning not based-on prior dose formore » H&N patients. Methods: Ten locally-advanced H&N cancer patients were selected for this study. For each patient, six weekly CT imaging were acquired during the course of radiotherapy. PTVs, parotids, cord, brainstem, and esophagus were contoured on both planning and six weekly CT images. ART with weekly re-plans were done by two strategies: 1) Generating a new optimized IMRT plan without including prior dose from previous fractions (NoPriorDose) and 2) Generating a new optimized IMRT plan based on the prior dose given from previous fractions (PriorDose). Deformable image registration was used to accumulate the dose distributions between planning and six weekly CT scans. The differences in accumulated doses for both strategies were evaluated using the DVH constraints for all structures. Results: On average, the differences in accumulated doses for PTV1, PTV2 and PTV3 for NoPriorDose and PriorDose strategies were <2%. The differences in Dmean to the cord and brainstem were within 3%. The esophagus Dmean was reduced by 2% using PriorDose. PriorDose strategy, however, reduced the left parotid D50 and Dmean by 15% and 14% respectively. Conclusion: This study demonstrated significant parotid sparing, potentially reducing xerostomia, by using ART with IMRT optimization based on prior dose for weekly re-planning of H&N cancer patients.« less