Adaptive radiation therapy (ART) using magnetic resonance (MR)-guided linear accelerators (MR-LINACs) is an emerging approach that allows for daily treatment plan adaptation. However, bulk density overrides (BDOs), used during adaptive planning (e.g., on the Elekta Unity system), may introduce dosimetric uncertainties compared to reference computed tomography (CT) images by approximating electron densities (ED). This study aims to evaluate and quantify the dosimetric differences between BDO-based plans and corresponding reference CT-based plans within the Monaco treatment planning system (TPS) for prostate MR guided adaptive radiotherapy (MRgART). Using a retrospective analysis approach, 21 anonymized patients undergoing prostate volumetric modulated arc therapy (VMAT) were selected. CT numbers from the planning CTs were converted to electron densities using the CT-to-ED table in use at our institution. Dose from the original plan was calculated on the CT dataset in the Monaco TPS (used as reference). Copies of each plan were then made, in which a BDO was applied by assigning a homogeneous Hounsfield Unit (HU) value to each delineated structure. Dose was recalculated with the overridden densities (converted to ED via the same CT-to-ED table) and compared to the reference plan by baseline and QUANTEC (Quantitative Analysis of Normal Tissue Effects in the Clinic) dose-volume histogram (DVH) metrics, expressed as a percentage of the prescription dose. Regions with increased density heterogeneity, such as air in the rectum, exhibited the largest differences, with some baseline DVH metrics (D50%) differing by up to 9%. No QUANTEC DVH metrics differed by more than 1%. The average of the mean dose difference to the prostate contour was 0.6%. No other contoured volumes of interest exceeded a 1.0% difference in mean dose. The findings of this study underscore the clinical viability of using BDOs for MRgART within the Elekta Unity / Monaco workflow. While there were localized dose discrepancies from the BDO, especially in the rectum, it did not affect the critical QUANTEC DVH metrics in any significant way. As this evaluation was performed on a single TPS platform, direct generalization to other systems (e.g., RayStation, Eclipse) should be made cautiously. Understanding the implications of BDO on dose distribution is crucial for optimizing treatment planning strategies and ensuring delivery of safe and efficacious radiation therapy in the context of MRgART.
We have trained and externally validated a knowledge-based planning model for radiation therapy planning in the setting of high-grade glioma. Model performance and utility in the context of clinical trial radiotherapy quality assurance (RTQA) are presented. A RapidPlan (RP) model was trained on 65 cases and tested on an additional 20 cases that were manually optimized and delivered within our institution. The model was externally validated on 34 cases that were manually optimized and treated at four outside institutions. These cases were selected to have target overlap with the brainstem or optic pathways. RapidPlan-generated plans were evaluated against planning objectives and manually optimized clinical plans. Cases were classified as (1) Clinical plan was superior, (2) Clinical plan was within 5 % of RP, or (3) Clinical plan could be improved. The clinical plan was characterized as superior for an objective if the metric was >5 % better than in the RP plan or if the clinical plan met the objective but the RP plan did not. Possible clinical plan improvement was indicated for an objective if the metric was >5 % better in the RP plan, or if the RP plan met the objective but the clinical plan did not. A Wilcoxon signed-rank test with a p < 0.05 significance threshold was used to determine if differences in PTV coverage, OAR doses and MU were statistically significant. Eight of 34 RP plans met all planning objectives in a single optimization, while an additional six met all normal tissue objectives while compromising target coverage. In more than 80 % of cases, when an objective was not achieved by RP, it was also not achieved in the manually optimized plan. Comparisons of clinical plans and RapidPlan indicated that statistically significant improvement was possible for Optic Nerve Dmax and Optics PRV V54; however, RP also introduced a statistically significant increase in Dmax overall. Clinical plans could have been improved for individual planning objectives 24 to 68 % of the time, while the clinical plans were considered superior to RP for individual planning objectives 9 to 51 % of the time. Improvement in PTV coverage was possible for 18 % of clinical plans. The presented RapidPlan model for high-grade glioma performs well for cases both within and outside our institution. The model has demonstrated the capacity to reduce normal tissue dose to optic structures and to provide feedback in the context of clinical trial plan RTQA.
In this study, we aimed to investigate the effects of different block combinations in helical tomotherapy (HT) on the dose distribution in the target volume, protection of the organs at risk (OARs), and treatment efficiency in patients with bilateral breast cancer (BBC), and to provide a dosimetric basis for the selection of individualized radiotherapy plans. Clinical data of 21 patients with BBC without distant metastasis, who underwent initial radiotherapy from January 2021 to September 2024, were retrospectively analyzed. Based on the properties of blocks-complete block (CB), directional block (DB), no block (NB) and their positions: spinal cord and bilateral lungs, 5 block combination plans were designed for each patient: SNLN (spinal NB/lung NB), SCLN (spinal CB/ lung NB), SCLC (spinal CB/lung CB), SNLC (spinal NB/lung CB), SDLD (spinal DB/lung DB). The dosimetric parameters for the target and OARs, treatment plan monitor units (MUs), and beam-on times (BOTs) were compared. All 5 plans met the clinical requirements for the target, V100%, V110%, homogeneity index, and conformity index with no statistically significant difference (p > 0.05). The lung CB significantly reduced the heart and lung irradiation doses. The spinal CB reduced the maximum dose to the spinal cord in the supraclavicular lymph node drainage area from 25 to 35 Gy to 4-9 Gy (p < 0.01). The lung CB group (Plans SCLC,SNLC) increased the BOT by 35% (p < 0.01) and MU by 2596 to 3060 compared to the lung NB group (Plans SNLN,SCLN). The DB (Plan SDLD) group balanced the OAR sparing and treatment efficiency; cardiopulmonary doses in the DB group were between those in the lung CB and NB groups (p < 0.01). The HT block technique significantly enhanced OAR protection. Lung and spinal CB maximized cardiopulmonary and spinal cord dose reductions at the expense of prolonging the BOT and MU. DB could be an alternative strategy for patients requiring a shorter BOT.
To build a knowledge-based treatment planning model using RapidPlan for esophageal cancer in the upper, mid and lower thoracic sites while comparing the performance of the model with and without the use of avoidance sectors. Plans from 45 patients with esophageal tumors requiring treatment in the upper (8), middle (15) and lower (22) thoracic regions were selected for training. Prescribed dose ranged from 41.4Gy to 56Gy (1.8-2 Gy/fraction) with simultaneous integrated boost (SIB). Clinical plans were re-planned without the use of avoidance sectors. Plans with reduced V105%(cc) to the planning target volume (PTV) while meeting criteria for organs at risk (OARs) were chosen to build the model. The model was validated on 23 plans by comparing the clinical plan to the RapidPlan with and without the use of avoidance sectors. RapidPlan without avoidance sectors reduced the V105%(cc) by 100.1 cc (p< 0.001), compared to the clinical plan, while also reducing the maximum dose. The lung V5 Gy was increased by 5.5% on average (p = 0.02). Doses to the heart, stomach, kidney, large bowel, small bowel and the spinal cord were not compromised. Using avoidance sectors with RapidPlan did not improve plan homogeneity or coverage over the clinical plan, while increasing the V20 Gy to the lungs by 2.8% (p = 0.01). Mean heart dose was increase by 1.1 Gy (p = 0.04), while kidney V18Gy (%) and spinal cord maximum were increased by 7.2% (p = 0.02) and 3.8 Gy respectively (p = 0.01). Volume of the prescription dose outside the PTV was lower by as much as 67 cc while using RapidPlan without avoidance sectors compared to the clinical plan and conformity index (CI) was improved by 7%-8% compared to the clinical plan. Using avoidance sectors worsened the CI by 3%-5% compared with the CP. The RapidPlan model configured for upper, mid and lower thoracic regions without the use of avoidance sectors significantly reduces volumes of the higher doses in the target and therefore potential likelihood of esophageal toxicity without excessively increasing lung dose.
To validate a script designed to streamline equivalent dose in 2 Gy fractions (EQD2) calculations for external beam radiotherapy. The VarianESAPI-EQD2Converter script for EQD2 calculation was evaluated for the Eclipse v. 16.1 treatment planning system. Seven anatomical sites were included (head and neck, lung, liver, breast, prostate, abdomen, and brain), and 3 cases per site were randomly selected from our clinical database. For each plan, the values of the D2%, D50% and D98% metrics of the planning target volume (PTV), and D2%, D50%, and D0.1cc of the organs-at-risk (OARs) were retrieved and converted into EQD2 values using the EQD2 formula ("true values"). Then, the script was executed for each plan following these 2 approaches: (1) the alpha-beta values of the PTV and OARs were simultaneously entered (approach 1); (2) only the external patient contour ("body") structure was considered, and the script was run several times by entering the alpha-beta values of the PTV and OARs each time. Validation of the script was performed by comparing the EQD2 values computed by the script for the Dx%/D0.1cc metrics with the respective true values. A 2-tailed t-test statistical analysis was done (alpha = 0.05). Nonstatistically significant differences were found between the 2 approaches, but unacceptably high differences (up to 57.5%) were found using approach 1. By contrast, low absolute differences (mostly < 0.5%) were found with approach 2. The VarianESAPI-EQD2Converter script with approach 2 is a reliable method to automate the EQD2 calculation process for Eclipse users.
Advancements in radiotherapy have led to diverse fractionation schemes for cancer treatment. Dosimetric assessment for moderately hypo-fractionated radiotherapy (mHFRT) plans is usually challenging due to the absence of explicit dose thresholds for these fractionations and the limitations of standard equivalent dose in 2 Gy per fraction (EQD2) conversion approach in dose volume histogram (DVH) conversion. This study aims to develop a method for converting physical DVHs of mHFRT into equivalent DVHs for conventional-fractionated radiotherapy (CFRT). This study presents an improved method, named EQDd (equivalent dose in d Gy fractions), for converting physical DVHs of mHFRT to equivalent DVHs of CFRT. Unlike the traditional EQD2 conversion method that uses a constant dose of 2 Gy per fraction, the EQDd method converts each mHFRT dose point with variable per-fraction doses. The conversion formula is expressed as EQDdi=Di*(1+di/(α/β))/(1+2*di/(dP*α/β))(i=1,2,…m), where dp is the prescription dose per fraction. Validation and illustration were performed by comparing the EQDd and EQD2 conversions for DVHs of an exemplary thoracic plan with 3 hypothetical prescriptions: (1) 2 Gy x 30; (2) 3 Gy x 20; (3) 4 Gy x 15, respectively. Additionally, a plugin for the Eclipse treatment planning system based on Eclipse Scripting Application Programming Interface (ESAPI) was developed to automate these conversions. For the 2 Gy x 30 prescription, the DVHs after EQDd conversion keep the same as the original DVHs, while EQD2 underestimates the volumetric doses of organs at risk (OARs). For the mHFRT plans (3 Gy x 20 and 4 Gy x 15), the DVHs after EQDd conversion show higher dose/volume than the original DVH for all structures, reflecting the increased biological effect for larger fraction sizes. While EQD2 appears to be only suitable for evaluating hotspots in the target and nearby serial organs, its utility for assessing full DVH patterns is limited. A plugin ``DVH Convert'' based on ESAPI has also been developed, which can convert DVH of an mHFRT to CFRT by EQDd method within just a few seconds. This study demonstrated the limitations of the EQD2 conversion method for OARs. The improved EQDd method provides more biologically reasonable conversions, which may have important applications for clinical dose evaluation.
In our department, monitor units (MU) for electron treatments are calculated using a Monte Carlo algorithm in the RayStation treatment planning system and independently checked using RadCalc software. An audit found that 22 of 50 independent checks failed the ± 5% local tolerance level. One of the causes for the discrepancy was thought to be that the RadCalc calculation takes no account of the patient surface curvature. Three-dimensional printed moulds were used to create wax "domes" and "bowls" with curvatures ranging from +0.33 cm-1 to -0.13 cm-1. The dose per MU delivered at dmax, under the wax phantoms, to a calibrated NACP type chamber was determined by measurement. The fields were delivered using 6 to 18 MeV electrons from a Varian TrueBeam linac. The measured data shows that the dose varies linearly, for a given electron energy, with the curvature of the wax mould. This information was used to create a set of curvature correction factors to be applied to the 50 independent MU checks. When the appropriate correction factor was applied to the RadCalc MU only 3 of the 50 calculations were beyond the tolerance level, improving agreement and reducing the number of out of tolerance results.
This study evaluated the effectiveness of an integrated Artificial Intelligence (AI) planning tool in a lung stereotactic ablative body radiotherapy (SABR) planning workflow. The aim was to determine whether the AI planning tool would facilitate the generation of consistent high-quality plans while simultaneously improving treatment plan efficiency. The study compares clinically treated planner derived lung SABR plans with AI-generated. Nineteen cases planned with traditional planner derived techniques which make up the control cohort human, were re-planned using AI to determine the efficiency and quality of AI generated plans. The study derived a set of AI criteria to create the AI cohort of plans, and further refinement with an additional optimization created AI + human cohort. Each plan was assessed using departmental criteria, including time efficiency, to determine plan quality. The best plans, chosen after a blind review by the treating RO, were documented and analyzed to demonstrate the effectiveness of AI assistance in Lung SABR planning. Ethics approval was given for this study at a local health district level. Across 19 patients, the human cohort showed a total of 3.3% criteria unmet, which dropped to 2.6% for AI assisted plans in the AI cohort. The percentage of unmet goals was further reduced to 1.84% after the addition of manual planner input in AI + human cohort. All plans selected by the RO in the blind review were produced using AI + human input, and the average time taken to produce AI assisted plans was 1.08 hours. The study demonstrates that AI, in conjunction with human expertise, significantly enhances the efficiency and quality of lung SABR plans for patients, with quality confirmed through blinded evaluation.
Linear accelerators (linac) with embedded magnetic resonance (MR) imaging involves a specific workflow that includes systematically a deformable image registration (DIR) with computed tomography images for the management of electron densities. Poor quality registration outputs could cause suboptimal dose calculations and planning treatments. The treatment planning system of the 0.35 T MR-linac MRIdian (ViewRay Inc., Oakwood Village) includes such a multimodal DIR tool with several adjustable parameters. To assist Viewray's users, this work investigates the influence of each parameter from the DIR tool. The default set of DIR parameters has been questioned through body locations (Thorax, Abdomen, Pelvis) and two quantitative metrics (the mutual information and mean absolute error). While the iterations number could be higher defined by default, the pyramidal number parameters (or control grid with a low impact) should be more carefully escalated. The default regularization parameters are consistent; both ways ensure optimal results, and the stiffness default value should be considered as a safeguard against unrealistic deformations. Depending on the location considered, the same variation in a specific parameter can have different effects on the registration results. Although the quantitative metrics are not directly correlated with the quality of the registrations, this work should help users manage and optimize the DIR step.
This study aimed to investigate the variable relative biological effectiveness with respect to the fixed RBE (1.1) on the prostate intensity-modulated-proton-therapy (IMPT), specifically focusing on normal tissue complication probabilities (NTCPs) in organs-at-risk (OARs) surrounding prostate tumors. The primary objective was to compare NTCP values between plans with fixed RBE (1.1) and plans with variable RBEs, considering the potential implications on treatment outcomes and normal tissue toxicity. Twelve prostate cancer patients undergoing simultaneous-integrated-boost treatments on a focal intraprostatic tumor (IPT) were studied. A linear function of the dose-averaged linear-energy-transfer (LETd) was utilized to generate variable RBEs. IMPT plans with fixed RBE (1.1) and variable RBEs were generated with the same beam and spot parameters. Dosimetric evaluations, including dose-volume histograms (DVHs) with RBE weighted doses, were performed to assess the impact of RBE variations on dose distribution. NTCP calculations were conducted using the Lyman-Kutcher-Burman (LKB) Probit model to quantify the risk of normal tissue complications in the bladder wall and rectal wall. The analysis revealed notable differences in NTCP values between plans with fixed and variable RBEs. The bladder wall and rectal wall showed increased NTCP values in plans with variable RBEs compared to fixed RBE plans. The average NTCP differences were 5.7% for the bladder wall and 4.6% for the rectal wall, with maximum differences reaching 9.5%. These variations were attributed to the doses lying in the steep region, where even slight dose increases resulted in significant NTCP elevations. Incorporating variable RBEs in proton therapy treatment planning leads to higher NTCP values in OARs, indicating an increased risk of normal tissue complications. This highlights the importance of accurately accounting for RBE variations to optimize treatment outcomes and minimize the potential for radiation-induced toxicities in surrounding organs.
Radiation therapy, a mainstay of management for prostate cancer, may lead to variations in rectal volume and subsequent toxicities. This study examines the role of bowel preparation compliance and Milk of Magnesia (MoM), on rectal diameter, toxicities, and volume variations during image-guided radiotherapy (IGRT). Through this, we aim to enhance treatment precision and reduce rectal toxicity, improving patient outcomes in pelvic irradiation. This prospective observational study, conducted at the Aga Khan University between July 2024 and December 2024, included prostate cancer patients receiving pelvic radiotherapy. Patients underwent a standardized bowel preparation protocol involving an antiflatulence diet and Milk of Magnesia (MoM) in an attempt to reduce rectal volume and toxicity, with bladder preparation requiring 500 mL of water intake 30 minutes before scans. Daily cone beam CT (CBCT) images ensured accurate target coverage, and rectal parameters were monitored throughout treatment. Data on patient compliance, rectal volume, interfractional motion, and toxicity were collected using the Adaptive Radiotherapy Integrated Algorithm (ARIA) system and self-designed questionnaires, with statistical analyses, including mixed-model logistic regression, assessing the relationship between various factors and treatment-related toxicity, defined as significant at p < 0.05. A total of 145 scans involving 7 prostate cancer patients undergoing radiotherapy were analyzed, with a mean patient age of 63.9 years and mean diagnostic serum PSA levels of 27.0 ng/ml. Most patients had a Gleason score of 5 and were treated primarily with Volumetric Modulated Arc Therapy (VMAT), receiving a radiation dose of 7800 cGy over 39 fractions. Compliance with bowel preparation protocols and Milk of Magnesia (MoM) was suboptimal (25.5%). Statistical analysis revealed a significant association between MoM compliance and reduced rectal toxicity, while dietary compliance, although beneficial, did not achieve significance in multivariable models. The mean rectal dose during treatment increased significantly, while individual variations in rectal diameter and volume were observed. This study emphasizes the importance of strict adherence to bowel preparation protocols and MoM in mitigating rectal toxicity and improving patient outcomes in pelvic irradiation for prostate cancer.
Volumetric modulated arc therapy (VMAT) is a common radiation treatment technique that delivers radiation to the planning target volume (PTV) while sparing the organs at risk (OAR). One limitation of VMAT on Varian linear accelerators is the limited distance the multileaf collimators (MLCs) can travel across the x-axis of the treatment field. The researchers in this case study sought to evaluate whether a split-field technique offers improved sparing of OAR, particularly dose to the femoral heads, in postmenopausal women undergoing treatment for the endometrium and pelvic lymph nodes. In the following retrospective study, 3 patients with endometrial cancer were selected. Two plans were generated for each patient, 1 split-field and 1 full-field. The full-field plan covered the full width of the PTV in both arcs. The split-field plan was created by reducing the X1 jaw to 2.0 cm on the clockwise arc, and the X2 jaw was reduced to 2.0 cm on the counterclockwise arc. PTV coverage along with OAR doses, was evaluated for all plans. All 3 patients showed lower doses to the femoral heads in the split-field plan. Doses to the rectum, bladder, and small bowel were also lower in the split-field plan. Differences in the dose were sometimes very small between the 2 techniques. Overall, the split-field technique demonstrated improved dose reduction over the full-field technique, but more research with a larger sample size is required to confirm the split-field technique.
The Octavius 4D QA device uses an inclinometer mounted on the linear accelerator (linac) gantry stand to maintain detector alignment during VMAT delivery. Due to linac acceleration, combined with accuracy limitations of the inclinometer, the Octavius system may record incorrect gantry angles, potentially affecting the accuracy of reconstructed dose distributions. The purpose of this study was to evaluate the angular accuracy of the Octavius. Inclinometer angles were corrected using 2 methods: interpolation of corrections for a set of static gantry angles (Table-corrected), and data extracted from the linac log files (Log-corrected). Corrections were then applied to PSQA measurements, and their impact on PSQA results was assessed. The Table-correction ranged from - 0.2° to 1.1°, while the Log-correction ranged from - 2.0° to + 2.8°. Lag values increased with gantry rotation velocity. Both short (50° to 100°) and long (> 100°) arcs showed similar lag magnitudes and distributions. Although Log-correction improved PSQA results for both short and long arcs (p < 0.001 for both), the effect was more pronounced for short arcs (r = 0.7) compared to long arcs (r = 0.4). The effect of Table-correction on PSQA was statistically significant for long arcs (p = 0.009) but weak (r = 0.3), and it did not significantly affect short arcs (p = 0.546, r = 0.09). The magnitude and effect of corrections on measurements was investigated and possible explanation for lower passing rates in short arc measurements was provided.
Ventricular tachycardia is a life-threatening cardiac arrhythmia that can potentially be treated using stereotactic body radiation therapy given as a single fraction. Given these characteristics, along with the delicate nature of irradiating a portion of the heart, accuracy of dose calculation in the treatment planning system is imperative. Two such calculation methods employed are superposition-convolution-based algorithms, such as Varian's Analytical Anisotropic Algorithm (AAA) and Linear Boltzmann Transport Equation solvers, such as Varian's Acuros XB (AXB). The purpose of this work is to compare these calculation algorithms in this treatment setting. Forty-eight patient plans with a nominal planning target volume prescription dose of 25 Gy that were initially calculated using AAA were recalculated using AXB without re-optimization. A third recalculation was conducted using Radformation's standalone Monte Carlo (MC) algorithm as a gold standard. Both AAA and AXB dose distributions were then compared to MC using 3D gamma analysis. Various target dose metrics in the AAA, AXB, and MC plans were compared along with organ-at-risk (OAR) dose points. Relative to MC, the median agreement rate for AAA was 89.1% and for AXB was 98.3%. The minimum agreement rate over the patient cohort was 70.4% and 93.2% for AAA and AXB, respectively. The overall gamma agreement rates were significantly higher for AXB than for AAA. With respect to the target and OAR dose comparisons, in every case, the differences in dose metrics were significantly different between AAA/MC and AXB/MC, with AXB having the superior agreement with MC. The mean differences between MC for both algorithms were lower in each case for AXB. By every metric evaluated, including target dose coverage, OAR dose endpoints, and 3D gamma dose evaluation, AXB resulted in a more precise agreement with the MC dose calculation benchmark for cardiac radioablation of ventricular tachycardia.
Testes exist as a highly radiosensitive organ at risk (OAR) in which absorbed dose can be significantly impacted by small positional variations. In this case study, challenges arose during radiation therapy treatment regarding immobilization and positional reproducibility of the patient's left testis and penis related to the body's homeostatic response to variable room temperatures. Researchers aimed to describe how the OAR constraints were maintained when room temperatures impacted scrotal positioning by comparing daily localization scans using computed tomography on rails (CToR) to evaluate the dose distribution. In this single pediatric case study, a patient diagnosed with paratesticular sarcoma received intensity modulated proton therapy (IMPT) treatment to the right inguinal region. Immobilization and positional reproducibility inconsistencies of the penis and left testis led to multiple replans. A retrospective statistical analysis of the OAR metrics in 3 radiation treatment plans determined specific patient warming instructions yielded a consistent reproducible outcome along with increased dosimetric accuracy. Understanding the human physiological response to ambient factors, such as room temperature, led to stable reproducibility of the OAR using thermal management. Iterative planning and collaboration between radiation therapy professionals were also key elements that paved the way to maintain treatment planning constraints.
Radiotherapy (RT) treatment planning requires the evaluation of multiple, often conflicting dosimetric and clinical parameters. Traditional tools such as dose-volume histograms and isodose maps continue to play a valuable role in clinical evaluation; however, these tools do not support the comparison of complex datasets. Therefore, evaluating the planning parameters may exceed the cognitive capacity of clinicians to interpret the information effectively. This study proposes a structured multicriteria decision-making (MCDM) framework to facilitate objective and systematic selection of optimal RT treatment plans, enhancing the consistency of clinical evaluations. Whole left breast RT plans were generated using 3 external RT modalities - 3-Dimensional Conformal Radiotherapy Field-in-Field, Intensity-Modulated Radiotherapy, and Volumetric Modulated Arc Therapy. Nine plans - 3 per modality, reflecting varying clinical quality levels: good, moderate and poor - were accomplished using an anthropomorphic female phantom. Evaluation criteria encompassed target volume dose-volume parameters, treatment plan parameters and organ-at-risk constraints. Objective weights were assigned using the CRITIC method. Five MCDM techniques were employed to rank the plans, and the results were aggregated using arithmetic mean and the COPELAND method. The rankings compared with a clinically derived reference ranking. COPRAS, ARAS, TOPSIS, and WASPAS showed strong agreement with the reference ranking, while GRA produced less consistent results. This study effectively demonstrates the feasibility of employing MCDM methods to provide a systematic, reliable, and data-driven framework for the selection of optimal radiotherapy treatment plans. The software developed for this purpose, designed to work with the Varian Eclipse Treatment Planning System, has the potential to support clinical decision-making by offering objective plan comparisons. Future studies should extend this approach by incorporating actual patient data instead of an anthropomorphic phantom, as well as by including different anatomical treatment sites to enhance clinical applicability and generalizability.
The advantages of intensity modulated radiotherapy (IMRT) in target coverage and organ at risk (OAR) sparing come with a low- and intermediate-dose bath effect. To control this unwanted dose dumping, planners use helping structures in the plan optimization. We hypothesized that using an anatomic definition of the posterior neck would result in reduced doses to the posterior neck muscles without compromising target coverage or increasing doses to organs at risk. We randomly selected twelve head and neck cancer patients treated with volume modulated arc therapy (VMAT). For each case, we generated 3 plans with different optimization methods: first, with a dummy volume behind the spinal vertebrae; second, a volume with the posterior neck and upper shoulder muscles according to anatomic criteria; and third, a plan without any helping structure. There was no statistically significant difference between the 3 plans in terms of coverage, conformity index or homogeneity index. All the OAR dose constraints were respected, and there was no significant increase of the dose to these organs when the anatomic definition for the posterior neck was used. However, the posterior neck muscles and upper shoulder muscles received significantly lower mean doses, 31.9 Gy vs 36.8 Gy vs 37.6 Gy, p = 0.0004. Using an anatomically defined posterior neck avoidance volume in the plan optimization significantly reduces doses to the posterior neck and upper shoulder muscles, potentially with less severe neck fibrosis, without compromising target volume coverage or increasing doses to other organs at risk.
This study aimed to assess intrafractional positional shifts occurring during cone-beam computed tomography (CBCT)-based online adaptive radiotherapy (oART) using the Varian Ethos system. Specifically, we quantified these positional deviations and examined their relevance for planning target volume (PTV) margin definition in the absence of pretreatment verification imaging. A retrospective analysis was performed on 28 patients who underwent oART, covering a total of 627 fractions. For each fraction, an initial CBCT was used for adaptive planning, followed by a second CBCT immediately before treatment to assess intrafractional motion. Shifts between the two CBCTs were measured in the anterior‒posterior (AP), superior‒inferior (SI), and medial‒lateral (ML) directions. The means ± standard deviations and ranges are reported. Additionally, CBCT acquisition intervals, total treatment times, and monitor unit (MU) differences between the reference and adaptive plans were evaluated. We found mean positional shifts of 0.11 ± 0.09 cm (AP), 0.07 ± 0.14 cm (SI), and 0.06 ± 0.09 cm (ML). We calculated PTV margins with all position corrections, ranging from 2.5 to 3.9 mm depending on the region and direction. CBCT intervals ranged from 11.4 to 17.5 minutes, and total treatment time from 14 to 45 minutes, depending on site. No strong correlation was observed between treatment time and the magnitude of positional shifts, although weak but statistically significant correlations were observed for prostate (AP) and rectum (ML). MU differences between the reference and adaptive plans reached statistical significance for prostate, endometrial, and bladder treatments. Intrafractional positional changes during oART are clinically relevant and may compromise geometric accuracy when not accounted for. In the absence of pretreatment imaging, individualized PTV margins based on treatment site and direction should be considered. Incorporating secondary CBCT or adopting site-specific margin strategies is recommended, particularly for pelvic treatments.
This study investigated hybrid volumetric-modulated arc therapy (H-VMAT) for the treatment of nonsmall cell lung cancer (NSCLC) by classifying patients based on the tumor location (left or right and upper, middle or low) and planning target volume (PTV) (less than average or greater than average) to determine the optimal VMAT dose ratio by dividing the prescription dose used for H-VMAT planning. The following treatment plans comprising four-field conformal irradiation were created for 51 patients with NSCLC: VMAT with one full arc (f-VMAT); VMAT with two partial arcs (p-VMAT); and hybrid plans. Hybrid plans comprised a combination of f-VMAT and three-dimensional conformal radiation therapy (3D-CRT; fH-VAMT) as well as a combination of p-VMAT and 3D-CRT (pH-VMAT). We evaluated the dose to the organ at risk using lung V5 and V20 Gy, heart V30 and V40 Gy, and the mean heart dose and determined the optimal dose ratio of H-VMAT. H-VMAT resulted in negligible differences in the PTV conformity and homogeneity of fH-VMAT and pH-VMAT regardless of the dose ratio. Compared to fH-VMAT, pH-VMAT significantly reduced low lung doses, particularly those for tumors located in the right and above the heart, and those for larger target volumes. Compared to fH-VMAT, pH-VMAT slightly reduced the heart volume exposed to medium or high doses. The optimal VMAT dose ratios based on the tumor conditions were as follows: 30 % fH-VMAT and 50 % pH-VMAT for left lung tumors; 20 % fH-VMAT and 50 % pH-VMAT for right lung tumors; 20 % fH-VMAT and 50 % pH-VMAT for tumors above the heart; and 50 % fH-VMAT and 50 % pH-VMAT for tumors at the heart level. For smaller tumors, both fH-VMAT and pH-VMAT were set at 50 %. For larger tumors, 20 % fH-VMAT and 50 % pH-VMAT were used. These findings can help optimize treatment planning and guidelines and contribute to improved patient outcomes.
The objective of this study was to evaluate the dosimetric characteristics and clinical utility of radiotherapy plans using the Dynamic Swing Arc (DSA) technique on the OXRAY (Hitachi Ltd., Tokyo, Japan) system for stage III non-small-cell lung cancer (NSCLC). This study involved a retrospective analysis of 26 patients with stage III NSCLC treated with non-stereotactic volumetric modulated arc therapy (VMAT). We generated DSA plans and conventional coplanar VMAT plans with identical optimization parameters. After contouring targets and organs at risk using three separate computed tomography images acquired under free breathing, inhalation, and exhalation conditions, the dose-volume parameters, including percent volume receiving ≥20 Gy (V20Gy), V5Gy, and mean lung dose for normal lungs, as well as spinal cord, heart, and esophagus doses, were evaluated. The parameters of total monitor units (MUs) and beam-on time (BoT) were also compared. Statistical significance was assessed using the Wilcoxon signed-rank test. Compared with conventional coplanar VMAT, the DSA plans significantly reduced normal lungs doses, with median V20Gy reducing from 19.18% to 16.46% (p < 0.001). DSA plans also reduced other lung dose metrics, including V5Gy and mean lung dose, while maintaining target coverage and homogeneity. A reduction in the median spinal cord (D0.03cc) of approximately 4 Gy in the DSA plan (p < 0.05), indicated greater sparing compared with other organs at risk. Total MU and BoT were higher in DSA plans (734.90 vs. 634.92 MU, respectively; 158.0 vs. 124.5 s, respectively; p < 0.001). However, the increase in BoT of several tens of seconds is small relative to the overall treatment time, including patient setup (approximately 10-15 min), and can be considered clinically acceptable. DSA plans on the OXRAY system for stage III NSCLC maintained target dose coverage and enabled significant sparing of normal lung tissue, including V20Gy. Despite modest increases in MU and BoT, clinical efficiency and accuracy were minimally affected. These results indicate that DSA may be a clinically valuable option for radiotherapy in stage III NSCLC with improved normal tissue protection.